A comparative study of root canal preparation using ProFile .04 and Lightspeed rotary Ni–Ti instruments

Materials and methods

Preparation of teeth
A modification of the Bramante technique (Bramante et al . 1987, Hülsmann et al . 1999b) was used to evaluate simultaneously the cleaning ability as well as preparation form (longitudinal and cross-sectional), safety issues, and working time on extracted teeth under conditions comparable to the clinical situation. A muffle-block was constructed, consisting of a u-formed middle section and two lateral walls which are fixed together with three screws. Grooves in the walls of the muffle-block allowed removal and exact repositioning of the complete toothblock or sectioned parts of the tooth. A modification of a radiographic platform, as described by Southard et al . (1987) and Sydney et al . (1991) could be adjusted to the outsides of the middle part of the muffle. This allowed the exposure of radiographs under standardized conditions and geometric relationship in order to allow the superimposition of views taken before, during and after root canal preparation. Two metallic reference objects inserted into the film holder facilitated exact superimposition of the radiographs. The system and the evaluation technique have been described previously (Hülsmann et al . 1999b).
Fifty extracted mandibular molars with two curved mesial root canals were opened and controlled for apical patency with a size 10 reamer. All teeth were shortened to a length of 19 mm. The teeth were mounted in the mould with acrylic resin and isolated with rubber dam and a clamp, simulating the clinical situation and ensuring the operator could gain access to the root canal only from the mesial direction. Root canal curvatures were measured as described by Schneider (1971) from preoperative radiographs after insertion of a size 15 reamer. The teeth were randomly divided into two groups. By exchanging a few single teeth a similar mean degree of curvature was achieved for both groups. Twenty-five teeth with 50 curved mesial root canals were prepared with the Profile .04 Ni–Ti system (Dentsply Maillefer), and 25 teeth were prepared with Lightspeed Ni–Ti rotary instruments (Lightspeed Inc.).

Instruments and preparation techniques
Profile .04
In the present study root canal preparation was performed in the following crown-down-sequence: Profile .04 size 25: 14 mm, size 30: 14 mm, size 20: 16 mm, size 15: working length (18 mm), sizes 20–45: working length (18 mm). The total number of instruments used was 10.

Lightspeed
Preparation with Lightspeed instruments was performed using a step-back technique (Wildey & Senia 1989, Wildey et al . 1992). The sequence of instruments used in the present study was the one proposed by the manufacturer: Hand instrument size 15: working length (18 mm), Lightspeed instruments sizes 20–45: working length (18 mm), sizes 47–70: step-back with each instrument used 1 mm shorter than the preceeding one. The total number of instruments (incl. size 15 hand-file) used was 20.
All root canals were prepared with a dental handpiece in a low-speed motor with torque-control (TCM 3000, Nouvag, Konstanz, Germany). Preparation speed was 350 r.p.m. for Profile .04 and 1300 r.p.m. for Lightspeed. Irrigation was performed with 2 mL NaOCl (3%) after each instrument size in the ProFile group and after each second instrument in the Lightspeed group with a final irrigation of 5 mL NaOCl (3%) in both groups. RC-Prep (Premier, Philadelphia, USA) was used as a chelating agent with each instrument. In both groups instruments were discarded after preparation of 10 root canals.

Assessment of preparation
First the mesio-buccal root canal was instrumented in the unsectioned teeth. Maintenance of root canal curvature, safety issues (loss of working length, apical blockage, instrument fracture, lateral perforation), and working time were evaluated at this time. Before preparation a radiograph with a size 15 stainless steel reamer in situ was taken and the initial root canal curvature was determined using the technique proposed by Schneider (1971). Following preparation to size 45 a further radiograph was taken with a size 40 stainless steel reamer. The outlines of the inserted instruments, the root outlines and the metallic reference objects in the film holder were superimposed under an X-ray viewer with a 10 magnification and the degree of straightening was evaluated by measuring the angle between the two instrument tips. The reference objects allowed control of exact superimposition of the radiographs.
The teeth were sectioned horizontally at 3, 6 and 9 mm from the apex and the preoperative root canal diameters were photographed under standardized conditions. The horizontal segments were remounted into the mould which was facilitated by the horizontal grooves and the mesio-lingual root canals were prepared to size 45 as described above. Again procedural accidents were recorded and straightening of the root canal curvature was measured using the radiographic platform. At the end of preparation the cross-section of the disto-lingual root canal was photographed again. According to Loushine et al . (1989) the postoperative cross-sections were classified as round, oval or irregular using reference photographs. Only irregular cross sections were regarded as unacceptable preparation results, since an oval cross section may be due to the cutting angle during the sectioning procedure. The divergence of pre- and postoperative root canal diameter was evaluated by superimposing pre- and postoperative canal outlines.
Following this the segments were removed from the mould and the three root segments were freed from the resin and split vertically. For the SEM investigation the mesiobuccal root canals, prepared before sectioning the teeth, were selected since irregular hydrodynamics in the sectioned roots could have influenced the degree of cleanliness. The buccal half of the split root canal segments was prepared for SEM investigation. The roots were coded and mixed so that the type of instrument used for preparation could not be identified during the SEM investigation.
Separate evaluations were undertaken for debris and smear layer with a five score system for each using the same set of reference photographs as in previous investigations (Hülsmann et al . 1997, 1998, 1999a, Hülsmann 2000, Hülsmann et al . 2001).
Debris was defined as dentine chips, pulp remnants and particles loosely attached to the root canal wall:

Score 1: Clean root canal wall, only few small debris particles
Score 2: Few small isles of debris
Score 3: Many accumulations of debris covering less than 50% of the root canal wall
Score 4: More than 50% of the root canal wall covered by debris
Score 5: Complete or nearly complete root canal wall covered by debris.
Scoring of debris was performed using 200 magnification.
Smear layer was defined as proposed by the American Association of Endodontists’ (1994) glossary Contemporary Terminology for Endodontics : A surface film of debris retained on dentine or other surfaces after instrumentation with either rotary instruments or endodontic files; consists of dentine particles, remnants of vital or necrotic pulp tissue, bacterial components and retained irrigant.

Score 1: No smear layer, dentinal tubules open
Score 2: Small amount of smear layer, some dentinal tubules open
Score 3: Homogeneous smear layer covering the root canal wall, only few dentinal tubules open
Score 4: Complete root canal wall covered by a homogeneous smear layer, no open dentinal tubules
Score 5: Heavy, inhomogeneous smear layer covering the complete root canal wall.
Smear layer was scored under 1000 magnification.
After the central beam of the SEM had been directed to the centre of the object by the SEM-operator (F.S.) under 10 magnification, the magnification was increased to 200 and 1000 , respectively, and the canal wall region appearing on the screen was scored. The scoring procedure was performed by a second operator (M.H.) who could not identify the coded specimens nor the device used for root canal preparation. This operator had been trained in the scoring procedure, resulting in a sufficient intraobserver reproducibility (Hülsmann et al . 1997).
The incidence of procedural accidents was assessed during preparation of both the unsectioned and sectioned root canals. Apical patency was controlled using an ISO 10 reamer extending 1 mm beyond working length.

Statistical analysis
Statistical analysis was performed using the following tests: for straightening Wilcoxon’s test was used ( P < 0.05); for comparison of the cross-sections and root canal cleanliness Fisher’s exact-test ( P < 0.05) was taken. The Mann–Whitney test ( P < 0.05) was used for comparison of working time.

Superimposition of photographs of the cross-sections of the pre- and postoperative root canals showed that both systems left uninstrumented canal walls in many cases. Overall, 40 Lightspeed and 30 Profile .04 specimens out of 75 showed no contact between the pre- and postoperative root canal outlines, indicating sufficient circumferential instrumentation of the root canal wall. Following preparation with Lightspeed instruments, all parts of the root canals showed less uninstrumented root canal walls than following preparation with ProFile instruments. Lightspeed instruments enlarged the root canal more uniformly with no specimen showing 50% or more contact between pre- and postoperative diameter (Table 3). The difference was statistically significant for the coronal third of the root canals ( P = 0.032) with the Lightspeed system showing a superior performance. For the apical and the middle third no significant differences occurred.

Root canal cleanliness.
Due to three instrument fractures, the number of specimens for the Profile .04 group was only 66 compared to 75 in the Lightspeed group. The results of the SEM analysis of the root canal walls concerning residual debris and smear layer are detailed in Table 4. Generally, the root canals showed no homogeneous appearance with only few specimens (Lightspeed: 14.7%, Profile .04: 13.6%) with completely clean walls without any remaining debris (score 1) and a high number of scores 2 and 3 for both systems (Lightspeed: 80%, Profile .04 71.2%). Differences between the systems were not significant.
For smear layer Lightspeed preparation resulted in 65.3%, ProFile in 62.1% scored 2 and 3. No statistically significant differences occurred for the apical and middle thirds of the root canals. For the coronal region, Lightspeed cleaned significantly better ( P = 0.029).

Procedural errors.
Lightspeed preparation proved to be a safe technique with no instrument fracture, no perforation, no apical blockage, and no case of loss of working length. With the Profile .04 system three instrument fractures (one fracture with size 40, two fractures with size 35 instruments) occurred, but no perforation, no apical blockage, and no cases with loss of working length.

Working time.
Working time, not including time for instrument changes and irrigation, measured during preparation of the unsectioned roots, resulted in a median of 123.4 s for Lightspeed instrumentation (20 instruments), and 94.3 s for the Profile .04 system (10 instruments). The difference was statistically significant ( P = 0.026). The working time for each single instrument was shorter for Lightspeed than for ProFile. Including time for root canal irrigation, Lightspeed preparation resulted in a working time of 270.9 s (median) and Profile .04 instrumentation in a median working time of 190.1 s.

Working safety.
High numbers of instrument fractures have been reported for Ni–Ti files in several earlier studies (Barthel et al . 1998, Kavanagh & Lumley 1998, Schäfer & Fritzenschaft 1999, Baumann & Roth 1999), indicating that Ni–Ti instruments may be more susceptible to separation than conventional stainless steel instruments. In the study of Barthel et al . (1998) four fractures occurred during the preparation of 10 root canals with curvatures of more than 35 . Schäfer & Fritzenschaft (1999) had nine fractures in 24 simulated root canals with a 35 curvature. Baumann & Roth (1999) reported that the incidence of fractures of Profile .04 increased with increasing size of the files with most fractures occurring with size 30 and 35 files. Thompson & Dummer (1998a) reported on three fractures and three deformed instruments after preparation of 40 simulated root canals to size 35.
Knowles et al . (1996) could complete enlargement of 20 root canals with Lightspeed to size 50 and Short et al . (1997) completed 15 curved canals to size 40 without fractures. Tharuni et al . (1996) reported on one fracture during preparation of 12 simulated curved canals. In the evaluation of Thompson & Dummer (1997c) Lightspeed proved to be a safe system without any fracture or apical blockage but some cases of loss of working length, which in no case exceeded 1 mm. In a survey amongst Swiss clinicians 76% of the respondents had fractured an instrument at least once. The main reasons for such fractures were non-constant speed of rotation, excessively high rotational speed, overuse of instruments, and too much pressure (Barbakow & Lutz 1997). As advocated by Gambarini (2000), the use of a newly developed low-torque endodontic motor with constant speed for Ni–Ti preparation with individually adjusted torque levels for each file of any Ni–Ti system instead of a hightorque motor might help to reduce the risk of instrument separation. Further studies could show that the frequency of instrument fractures may be related to the operator’s experience with a certain system (Glosson et al . 1995, Pruett et al . 1997, Barbakow & Lutz 1997, Baumann & Roth 1999).
The results of the present study are in general agreement with the results of the studies performed by Thompson & Dummer (1997a,b,c,d) in simulated root canals with different types of curvature. They described a very low incidence of procedural errors such as fractures, apical blockages, elbows, zips, or loss of working length for Lightspeed, as well as for Profile .04.

Working time.
The finding that ProFile instrumentation took significantly less time than Lightspeed preparation, to a large extent, will be due to the fact that the number of instruments for both systems differs (Lightspeed: 20, ProFile: 10). Following the protocol of this study the number of irrigations for ProFile was 10, for Lightspeed 12. Clinically, the difference will be even more evident as time for instrument changes and adjustment of the stopper systems has to be added. On the other hand, the preparation time for each single instrument was shorter for Lightspeed, due to the reduced contact zone between instrument and root canal wall. Conversely, Thompson & Dummer (1997a,b,c,d), in their studies on simulated root canals with different kinds of curvatures, could find nearly equal preparation times for Lightspeed and Profile .04, but when using Profile .04 with ISO-sized tips the working time was shortened substantially (Bryant et al . 1998a). These differences might be explained by the fact that the time an operator keeps the instrument working inside the root canal and the speed for the up and down movement of the files is not clearly defined and will vary individually. The total time for preparation for both systems was clearly shorter in the present investigation than in the studies mentioned above, as time for instrument change, stopper adjustment and irrigation were not included.
Overall, the ability of Ni–Ti systems to shorten preparation time compared to hand instrumentation or to automated root canal preparation using different endodontic handpieces with conventional stainless steel files has been confirmed already in a number of previous studies (Esposito & Cunningham 1995, Glosson et al . 1995, Hülsmann 2000, Hülsmann et al . 2001).