B. T. Tan & H. H. Messer
Department of Restorative Dentistry, School of Dental Science, University of Melbourne, Melbourne, Australia.

Introduction.
Effective canal debridement relies on the accurate determination of the working length (WL) and adequate apical canal enlargement. The extent of apical enlargement is typically based on anestimate of the initial canal size as determined by the size of hand file that ‘binds’at the WL (Grossman et al. 1988). Both tactile detection of the apical constriction and apical file size determination depend on the assumption that the canal is narrowest in the apical region, with unrestricted passage of the file to that point. Continued dentine formation is responsible for an increased thickness of dentine at the ￡oor of the pulp chamber and for progressive constriction of the canal space (Philippas 1961). This coronal constriction should be removed by preflaring for accurate WL and file size determination (Leeb 1983, Stabholz et al.1995, Contreras et al. 2001).
In the past, few authors have conducted clinical research on the ability to detect the apical constriction by tactile sensation, and the apical constriction is indeed considered a ‘myth’ by some authors. Seidberg et al. (1975) reported that the apical constriction could be identified accurately in 64% of cases with digital-tactile means as compared with 48% using an (early model) apex locator. Another study (Stabholz et al.1995) showed that determination of the apical constriction by tactile sensation was possible in 75% of cases if the root canal was preflared as compared to only 32.3% of cases if the coronal aspect was not preflared. The most significant finding in that study was that preflaring of the coronal portion of the root canal resulted in more reliable WL determination, implying that coronal constriction affected tactile discrimination in the apical part of the canal.
Leeb (1983) studied the effect of enlarging the rootcanal orifice on biomechanical canal preparation. Using India ink as a marker, he observed that normal dentine apposition caused the cervical region to be the narrowest portion of the root canal. If a Gates-Glidden drill or Peezo reamer was used to enlarge the orifice and to eliminate cervical interference, larger ￠les could be passed more easily to the apical constriction. Recently, Contreras et al. (2001) reported that early coronal flaring resulted in a significantly larger hand file (Flex-R file) fitting to the apex.
Lightspeed1(LS) is a non tapered rotary instrument. It consists of standard and half sizes having a cutting head approximately 1-2-mm long with a non cutting pilot tip. LS instruments have been reported to give a greater apparent apical size than conventionally tapered K-files (Levin et al. 1999, Liu & Jou 1999). A mean difference of 11.03 ISO units was found without preflaring the coronal part of the canal (Liu & Jou 1999). After ￡flaring, the mean difference was 7.25 ISO units. How ever, no further information was provided regarding the sizes of both instruments and whether only K-files or both instruments increased in sizes after flaring. Liu & Jou (1999) concluded that hand-held LS rotary instruments were better instruments to estimate the size of the apical constricture than K-files.
The above studies have demonstrated the importance of coronal flaring and the effect of different types of instruments on file size and vWL determination. If the Grossman criterion (Grossman et al.1988) of enlarging a root canal to at least three sizes beyond the first file that binds at WL is valid, then one should question whether a standard master apical file size of 25 or 30 would be su⁄cient for the apical preparation of narrow canals, as is routinely recommended by most authors (Grossman et al. 1988, Weine 1989, Ingle et al. 1994, Torabinejad 1994, Walton & Rivera 1996, West & Roane 1998). It may be more accurate to size each canal individually and subsequently determine its master apical file size to ensure that the apical third region is adequately enlarged and debrided prior to obturation.
The aim of this study was to investigate the effect of instrument type (K-files and LS instruments) and preflaring on the determination of initial apical size in a range of different canal types of varying sizes and curvatures. This study focused on posterior teeth only (premolars and molars) because these teeth impose the greatest challenge in root-canal treatment. This information is clinically important, as the extent of apical shaping will rely on the initial assessment of canal size.

Tooth selection.
Intact extracted maxillary and mandibular permanent premolars and molars (total:60 teeth) displaying normal pulp chamber, patent root canals and fully formed apices were selected (Table 1).Teeth with complicated anatomy, external root resorption or extreme root curvature were excluded from the study.
Standard access cavities were cut; the pulp tissues were removed with an extra-fine (XF) barbed broach (Antaeos CC-cord, Munchen, West Germany). Care was taken to ensure that the barbed broach engaged only the pulp tissue without contacting the apical third of the root canal. Canals were then irrigated with copious 1% sodium hypochlorite solution (Milton solution1, Procter and Gamble Australia, Parramatta, Australia). Apical patency was determined by inserting a size 6 K- file into the root canal until the tip of the file was visible at the apical foramen; the WL was set 0.5 mm short of the apical foramen. Cusp tips were used as reference points. Both the WL and the reference point of each individual canal were recorded. All the canals in each tooth (except the second mesio-buccal canals of maxillary molars) were included in the study. A total of121canals were eventually utilized. Each tooth was stored in an individually labelled, capped plastic vial containing 10% buffered formalin.

Table 1. Distribution of teeth selected for sizing.

Sizing of canals.
Each canal was sized four times using two instrument types before and after coronal flaring. Each canal was sized with both stainless steel K-files (Mani, Matsutani Seisakusho Co., Takanezawa-Machi Tochibi-Ken, Japan) and hand-held LS rotary instruments (Lightspeed Technology Inc., San Antonio, TX, USA) in random order. Files were inserted passively into the canal with a light ‘watch-winding’ action, and care was taken to avoid any force during sizing. For K-files, measurement was undertaken starting from ISO size 8, whereas for LS instruments, measurement began with size 20 (i.e. the smallest file).The biggest file size that reached the correct WL was recorded. The same procedure was used for each canal before and after coronal flaring.
The following precautions were taken to reduce systematic bias during measurement:

1. K-files and LS instruments were alternated as the first instrument used to size canals.
2. Sizing with the second file type was conducted in random order of teeth, without knowledge of the previous measurement result.
3. Each measurement was recorded on a separate sheet to prevent the operator from knowing the previous reading.

The only information the operator had each time before sizing a canal was the WL and its corresponding reference point.

Coronal and middle third flaring.
Coronal flaring and middle third flaring were done with Profile instruments (Maillefer-Dentsply, Ballaigues, Switzerland) using a crown-down approach to eliminate all interferences, cervical to the apical third region. flaring was terminated 4 mm short of the WL so that the apical third region remained unprepared. Recommended guidelines by the manufacturer (Profile 0.06/0.04 series) were adopted. Their sequence was as follows: Profile orifice shaper 3 (#40,0.06);  Profile orifice shaper 2 (#30, 0.06);  Profile #25, 0.06;  Profile #20, 0.06; Profile #25, 0.04; and  Profile #20, 0.04. Instruments were used sequentially until a point 4 mm short of the WL was reached. The final size used depended on the size of each canal. The typical instruments used up to 4 mm from the WL were  Profile orifice shaper 2 (#30,0.06) for premolar canals,  Profile#25,0.06 for moderate size mesial or distal molar canals, and Profile#25,0.04 for narrow mesial canals.

Statistical analysis.
A univariate analysis of variance (anova) was conducted to examine the effect of instrument type and flaring on the apical file size determination (dependent variable: apical file size; independent variables: instrument type, flaring). The interaction between instrument type and flaring was also examined. Post hoc tests (pair wise comparisons) were conducted only if there was a significant interaction between the two independent variables. All statistical analyses were performed at the 0.05 level of significance.

A total of 121 canals were utilized for analysis. Canal sizes are expressed either in ISO sizes or in diameters (x10_2 mm) (e.g. an ISO size 10 is equal to a file tip diameter of 10 x10_2 mm); file sizes typically increase in increments of 5 ISO units, or 5 _10_2 mm. The mean _ standard deviation diameters registered were as follows (Table 2) - K( without flaring): 23.8 _7.1; LS (without flaring): 33.4 _8.6; K (with flaring): 29.2 _ 7.3; LS (with flaring): 38.5 _8.5.
Both the instrument type and flaring had a highly significant effect on apical size estimate (P < 0.001,anova). The mean diameter of canal measurement with the LS instruments was larger than with K-files by 9.4 _ 10_2 mm (P < 0.001; 95% confidence interval: 8.7 _ 10_2,10.2 _10_2). flaring hadan impact on apical sizing by both types of instruments. It resulted in an increase of average diameter of 5.3 _10_2 mm(P < 0.001;95% confidence interval: 4.5 _10_2, 6.0 _10_2). There was no interaction between instrument types and flaring (P > 0.05), which means that the difference between K- files and LS was consistent before and after flaring, and the effect of flaring was consistent in both instruments (Table 3). Hence, no post hoc tests (pair wise comparisons) were conducted. Also, linear regression analysis did not show a significant relationship between the initial canal size (K-file) and the difference (LS _ K) (r2 ј0.028, P > 0.05), indicating that the differences between K-files and LS instrument readings were consistent as canal size increases.

Table 2. Range of canal size and mean diameter (x 10_2 mm) as registered by K-files and LS instruments before and after flaring (data for 121 canals).



Table 3. Range and mean diameter difference (x10_2 mm) between K-files and LS instruments used to determine apical canal size (data for 121 canals).

Effect of instrument type.
LS versus K-files Both before and after flaring, more than 90% of canals showed greater LS readings than the corresponding K- file readings (109/121 and 110/121 canals, respectively) (Figs 1and 2). The most common size difference (i.e. the mode) was 10 ISO units before flaring and 7.5 ISO units after flaring. The mean diameter differences of LS _ K were 9.6 _10_2 mm before flaring and 9.3 _10_2 mm after flaring, or approximately two file sizes (10 ISO units) larger (Table 3).

Effect of flaring.

After flaring versus before flaring.
Flaring resulted in an increase of the file size reading for both K-files and LS instruments. A large majority of the canals with K-file (77.7%) and LS (61.9%) registered at least one file size larger after flaring. The mean diameter differences after flaring versus before flaring were 5.4 _10_2 mm for K-file and 5.1 _10_2 mm for LS instruments, or approximately one file size (5 ISO units) larger (Table 3).

Figure 1. Scatter plot of apical file sizes registered for LS versus K-file (without flaring) (total:121canals). Points above the line indicate a larger LS size; points below the line indicate larger K-file size. Note that many of the points are superimposed so that all the 121 points are not visible.



Figure 2. Scatter plot of apical file sizes registered for LS versus K-file with flaring (total: 121 canals). Points above the line indicate a larger LS size; points below the line indicate larger K-file size. Note that many of the points are superimposed.

Discussion.
By removing the cervical interference, it was possible to insert a larger file size to the apical constricture. Hence, our present study has confirmed previous findings (Leeb 1983, Contreras et al.2001).The need for a larger file size to attain binding for both LS and K-file after coronal flaring is a further indication of the presence of cervical or mid root interference that initially prevents both types of instruments from reaching the apex. However, the ability of LS instruments to register a larger measurement than the K-file in the same canal may be owing to differences in instrument design and alloy properties. The short cutting blade, noncutting tip design and nontapered shaft of LS instruments (Fig. 3) may allow them to bypass the cervical constriction area or any premature binding site along the root canal. Furthermore, the flexible nickel-titanium alloy of LS instruments may allow them to bypass the canal curvatures more readily than a more rigid stainless steel file. Both the canal interference and the curvature may be the factors that govern a clinician’s ability to sense apical diameter with a file.
As in a previous study (Liu & Jou1999), the LS instruments registered larger apical file sizes than the K-files. However, the mean discrepancy between the two file types was unchanged after flaring (9.6 _10_2 mm vs. 9.3 _10_2 mm) rather than reduced (11 _10_2 mm vs. 7.3 _10_2 mm) (Liu & Jou 1999). In retrospect, one would expect the discrepancy between the two instruments to diminish after coronal flaring, as removing the cervical interference should allow a K-file to reach the apical 4 mm without obstruction. Possible explanations are a lack of canal taper (<0.02 mm mm_1 taper) and constrictions within the apical 4 mm, which allowed similar-sized LS but not K-files to pass through. Also large K-files may be too stiff to negotiate severe apical curvatures.
Two important issues need to be addressed in relation to the discrepancy of results obtained between the two instruments before and after flaring. Firstly, it is possible that repeated ‘sizing’ enlarged the canal even though the files were supposedly placed passively. Secondly, LS is not an instrument of any one determined shape that changes only in diameter. Rather, it is a series of instruments that show gradual shifts in both size and shape as the instrument size increases (Marsicovetere et al. 1996). Hence, part of the difference is possibly related to LS tip configuration because the maximum diameter of LS instruments may be further away from the pilot tip of the instrument than is the case for K-files. Conflicting reports (Marsicovetere et al. 1996, Schrader et al. 1998) suggest that LS instruments are undersized or oversized in relation to their nominal size. We measured the cutting heads of 26 LS instruments (from sizes 20- 50, two instruments of each size) using a micrometer. On average, instruments were slightly smaller than the stated size by approximately 1 ISO unit (0.01mm) and the difference varied from 0.5 to _3 ISO units. Thus, the results cannot be explained by size differences. Non etheless, it may not be appropriate to state that LS is more ‘accurate’ (i.e. closer to true canal diameter) than K-files in gauging the apical constricture.

Figure 3. A size 30 K-file with a corresponding size 30 LS instrument showing their different tip designs and taper (original magnification: 16x).

In the K-file group, the file size increased by approximately one file size after flaring rather than two file sizes as previously reported (Contreras et al. 2001). These differences may be due to variations in the types of teeth utilized in the two studies. Contreras et al. (2001) assessed only the mesial roots of mandibular molars, whereas the present study included a range of different canal types of varying size and curvature from both maxillary and mandibular premolars and molars. The values calculated are valid for ‘wider’canals, as the mean apical diameter of canals utilized was mostly file size 20 or 25 (i.e. K (without flaring), Table 2). Clinically, smaller canals may not be negotiable initially with a size 20 LS instrument.
In our study, 27 out of the 36 (i.e.75%) premolar canals sized by K-file recorded at least1 ISO file size larger after flaring. Furthermore, 31 out of 36 (i.e. 86%) premolar canals in the LS group registered larger file sizes after coronal flaring. This indicates that cervical interference is also present in premolar canals, and its removal will allow a larger file size to t to the apex. As a result, we would recommend early flaring not only in molar teeth but also in premolars.
Grossman et al. (1988) recommended that a root canal should be enlarged to at least three sizes beyond the first file size that binds at the apex to ensure adequate debridement. In our study, apical canal diameter determined with LS after flaring was approximately 3 ISO file sizes larger than with a K-file without coronal flaring. In other words, early flaring is essential and canals may need to be enlarged to a greater file size than previously accepted.
The LS manufacturer recommends early coronal flaring followed by individual sizing of each canal before determining the appropriate master apical rotary size. This usually results in a larger master apical file size. The typical master apical rotary sizes of 50-55 are recommended by the LS manufacturer (Lightspeed Technology Inc.) for the mesial canals of mandibular molars. These sizes are much larger than the ‘conventional’ master apical file sizes of 25-30 recommended by most authors for hand instrumentation techniques (Grossman et al. 1988, Weine 1989, Ingle et al. 1994, Torabinejad 1994,Walton & Rivera 1996,West & Roane 1998). Enlargement with stainless steel hand instruments larger than size 30 in mesial canals of mandibular molar will most likely lead to a high frequency of procedural errors, such as fledging, canal straightening, zipping, apical transportation and strip perforations. This is because most mesial roots of mandibular molars demonstrate curvature in both mesio-distal and bucco-lingual directions (Cunningham & Senia1992), and the dentine thickness is narrowest in the apical third region (Gani & Visvisian1999).
Finally, based on the results of our study and previously reported findings (Seidberg et al. 1975, Leeb 1983, Stabholz et al. 1995, Levin et al. 1999, Liu & Jou 1999, Contreras et al.2001), we conclude that if an operator wishes to determine an accurate canal WL and apical file size simultaneously, canal orifice enlargement should be performed first before placement of the measurement file. The better sense of apical diameter provides information that should result in better control of the biomechanical preparation. Furthermore, the use of LS instruments would register larger canal sizes than K-files in the apical constricture.
Early coronal flaring not only allows better sense of apical constricture and diameter, it may also facilitate cleaning by allowing the irrigant to work deeper, more quickly and more effectively into the apical third region (Ram 1977). The concept of apical enlargement is still poorly understood. To date, no study has shown the influence of apical enlargement on the success and failure of endodontic treatment. The feasibility of larger apical third preparation, especially in premolar and molar teeth, should be investigated. Questions such as whether the canals are cleaner with larger preparation (from both pulp remnants and dentine debris) and whether the roots are weaker if they are further enlarged, still remain to be answered.

References.

Contreras MAL, Zinman EH, Kaplan SK (2001) Comparison of the   first file that fits at the apex, before and after early flaring. Journal   of
  Endodontics 27, 113-6.
  Cunningham CJ, Senia ES (1992) A three-dimensional study of canal curvatures   in the mesial roots of mandibular molars. Journal of
  Endodontics 18, 294-300.
  Gani O, Visvisian C (1999) Apical canal diameter in the firs tupper molar at   various ages. Journal of Endodontics 25, 689-91.
  Grossman LI, Oliet S, Del Rio CE (1988) Preparation of the root canal: equipment   and technique for cleaning, shaping and irrigation. In:Grossman LI, Oliet S,   Endodontic Practice,11th edn. Philadelphia: Lea & Febiger, pp. 179-227.  
  Ingle JI, Bakland LK, Peters DL, Buchanan S, Mullaney TP (1994) Endodontic cavity   preparation. In: Ingle JI, Bakland LK, eds. Endodontics, 4th edn. Philadelphia:   Lea & Febiger, pp.92-227.
  Leeb J (1983) Canal orifice enlargement as related to biomechanical preparation.   Journal of Endodontics 9, 463-70.
  Levin JA, Liu DT, Jou YT (1999) The accuracy of two clinical techniques to determine   the size of the apical foramen. Journal of Endodontics 25, 294 (abstract - OR   47).
  Lightspeed Technology Inc. (2001) Lightspeed Technique Guide (for use with Simplifill1),   Lightspeed Product Information. San Antonio, TX, USA: Lightspeed Technology   Inc., http:// www.lightspeedusa.com/techniqueguide.htm.
  Liu DT, Jou YT (1999) A technique estimating apical constriction with K-files   and NT Lightspeed rotary instruments. Journal of Endodontics 25, 306(abstract   - PR 31).
  Marsicovetere ES, Clement DJ, Del Rio CE (1996) Morphometric video analysis   of the engine-driven nickel-titanium Lightspeed instrument system. International   Endodontic Journal 5, 231-5.
  Philippas GG (1961) Influence of occlusal wear and age on formation of dentin   and size of pulp chamber. Journal of Dental Research 40,1186-98.
  Ram Z (1977) Effectiveness of root canal irrigation. Oral Surgery, Oral Medicine   and Oral Pathology 44, 306-12.
  Schrader C, Sener B, Barbakow F (1998) Evaluating the sizes of Lightspeed instruments.   International Endodontic Journal 31, 295-300.
  Seidberg BH, Alibrandi BV, FineH, Logne B (1975) Clinical investigation of measuring   working lengths of root canals with an electronic device and with digital tactile   sense. Journal of American Dental Association 90,379-87.
  Stabholz A, Rotstein I, Torabinejad M (1995) Effect of preflaring on tactile   detection of the apical constriction. Journal of Endodontics 21,92-4.
  Torabinejad M (1994) Passive step-back technique. Oral Surgery, Oral Medicine   and Oral Pathology 77,398-401.
  Walton RE, Rivera EM (1996) Cleaning and shaping. In: Walton RE, Torabinejad   M, eds. Principles and Practice of Endodontics, 2nd edn. Philadelphia: W.B.   Saunders, pp. 201-33.
  Weine FS (1989) Intracanal treatment procedures, basic and advanced topics.   In: Weine FS, ed. Endodontic Therapy, 4th edn. St Louis: CV Mosby, pp. 277-369.  
  West JD, Roane JB (1998) Cleaning and shaping the root canal system. In: Cohen   S, Burns RC, eds. Pathways of the Pulp,7th edn. St Louis: CV Mosby, pp. 203-57.