Separated File Removal

INTRODUCTION
Ni-Ti files break mostly in narrow canals in the apical one third or curved canals because of the superelastic property.^1,2 As new Ni-Ti files become more elastic, ironically they break in more hard-to-reach areas of the canals which make the file removal process even more difficult.

Background
The use of ultrasonic devices has been shown to be very successful for the removal of various canal obstructions.^3-5 The use of ultrasonics is advantageous over conventional methods for removal of separated instruments as it is able to set separated instruments free from canals while making the sacrifice of dentin significantly less.⁶ However, there are several negative factors when using ultrasonics for the removal of separated files.

1a. Reseparated fragments done by ultrasonics.	1b. Retrieved fractured fragments.
2a. Separated file in the apical third of the canal.	2b. Separated file pushed out of the canal by ultrasonics.

First, ultrasonics sometimes causes secondary fracture of separated files, as shown in Figures 1a and 1b. Ward et al,⁷ in a study using ultrasonics to remove separated files from simulated canals and extracted teeth, observed that a portion of the separated instrument would occasionally break off from the original fragment, leaving a shorter fragment behind. Smaller fragments are more difficult to remove than larger fragments, and the removal rates are low for fragments that are located apical to the canal curvature.^8-10 Secondly, ultrasonic vibration sometimes pushes the separated file out of the canal, as shown in Figures 2a and 2b. Thirdly, aggressive use of ultrasonics sometimes results in perforating a canal, as shown in Figure 3. Fourthly, the prolonged use of ultrasonics can result in a temperature rise on the root surface.^11-13 If the temperature increase on the root surface goes beyond 10°C, the periodontal tissues can be seriously damaged.¹⁴ This is especially significant if ultrasonic energy is used without coolant to enhance visualization.^3,7,10,15,16 The increase in temperature generated by ultrasonics within the canal might be several times higher than the readings recorded on the external root surface.¹⁷ Therefore, if the temperature on the external root surface reaches 46°C, the temperature inside the canal can be approximately 200°C. The temperature rise depends on lengths of use, type of ultrasonic tips, power setting, and the use of coolant such as water and air coolant.^8,18 In fact, a recent study showed ultrasonic activation with constant air flow as coolant did keep the temperature rise significantly lower than that without the air flow during file removal attempts.¹⁷

Figure 3. File insertion into the perforated canal.

There were 2 studies conducted by the author on separated file removal.^19,20 One was done to investigate how secondary fracture occurred, and the other one was a retrospective study carried out to assess how the lengths of separated files influenced ultrasonic removal time. The results suggested that separated files longer than 4.5 mm in any canals, or those in curved canals greater than 60°, were extremely difficult to retrieve with ultrasonics alone; it took longer than 9 minutes to remove them, while separated files shorter than 4.5 mm long were easy to retrieve with ultrasonics alone and the removal time was within the range of 5 minutes. Therefore, in conclusion, ultrasonic removal attempts should be performed on a separated file shorter than 4.5 mm long, and ultrasonics should be replaced with something different, such as the loop device,^21,22 if ultrasonic removal time exceeds 5 minutes.
The other study disclosed that ultrasonic application to a separated file in contact with dentin wall behind made it significantly more resistant to secondary fracture than without dentin wall behind it. In fact, separated files in the curved canals did transfer more ultrasonic energy than those in the straight canals to the dentin wall that was behind the files to which ultrasonics were applied. Separated files with no surrounding dentin wall in the coronal portion were least resistant to secondary fracture by ultrasonics. It was recommended that, initially, ultrasonic application be limited to the canal wall coronal to the fragment and then to the inside of the curve during file removal attempts to avoid secondary fracture and file extrusion beyond apical foramen.

Instrument Removal Method
First of all, a straight-line access and canal enlargement to the separated file should be created with Nos. 2 and 3 Gates Glidden (GG) burs. This should be accomplished with minimal removal of the dentin to conserve the root structure, while the canal should be enlarged to at least 4 sizes (0.2 mm) larger in diameter than the separated file (Figures 4a to 4d). If the canal were shaped to larger than 1.0 mm in diameter during the file removal attempts, the chances of stripped perforation, especially in curved canals, would increase. Because the diameter of the No. 3 GG bur is 0.9 mm, it is the maximum GG bur to be used for this purpose of canal preparation. The GG bur has a pilot tip that follows the path already created by the separated file before the breakage. The preparation for the straight-line access usually takes less than 10 seconds with the GG bur.

Figures 4a to 4l. Recommended method to remove a separated file from a curved canal. 4a. Original canal with a separated file. 4b. Straight-line access with No. 2 Gates Glidden (GG) bur. 4c. Straight-line access and enlargement to the separated file with No. 3 GG bur. 4d. Canal status after the straight access and canal enlargement were established. 4e. Straight access to the fragment has to be created. 4f. Create a pocket between the file and the inside wall of the curve. 4g. Shallow groove is cut along the outer curvature. 4h. Fill the canal with EDTA solution to cool down the heat generated during ultrasonic vibration. 4i. Activate the ultrasonic tip in the slit and move the tip in “push and pull” motions. 4j. Ultrasonic vibration and acoustic streaming helps free the fragment file from the canal. Removal of the broken file completed. 4k. Loop is placed over the fragment. 4l. Loop is pulled in all directions.

A prototype ultrasonic tip is brought into the canal and then it is activated on the dentin wall of the inner curvature to create a tiny pocket approximately 1.0 mm deep from the severed surface of the file fragment (Figure 4e and 4f). Once this narrow space is obtained, a shallow groove is cut along the outer curvature so that there will be no obstruction that keeps the fragment from being kicked coronally out by ultrasonic vibration (Figure 4g). Then, the canal is filled with EDTA solution to enhance the ultrasonic cavitation effect and acoustic streaming for removal (Figure 4h).
Ultrasonic vibration should be applied to the separated file in the space created between the fragment and the inner curve of the canal, and move in “push and pull” motions until it is removed (Figure 4i and 4j). Most separated files (from my experience) usually come out in about 10 seconds with ultrasonics. However, if the separated file shows resistance to disengagement for more than 60 seconds, more dentin needs to be removed apically along the inside wall of the fragment in the hope of loosening it from the canal walls. Deepening the pocket of dentin wall with ultrasonics along the fragment most likely wraps up file removal attempts. If the separated file shows no sign of disengagement after the attempts for 5 minutes, the loop device will be used to capture the coronal portion of the separated file to pull out of the canal. Once the fragment is peripherally exposed by at least 0.7 mm, the loop is placed over the coronal portion of the separated file and then fastened to secure the fragment (Figure 4k). The obstruction is retrieved by pulling the loop out of the canal in various directions to dislodge the fragment from the canal walls (Figure 4l).

CASE 1
A 33-year-old female, referred to a private endodontic office, was seeking retreatment of her mandibular right first molar. The patient was in good general health with no significant past or present illness. Her chief complaint was intermittent pain while chewing ever since the tooth was treated 2 years ago. Clinical examination of the tooth revealed slight sensitivity, a healthy periodontium with no mobility.
The tooth was computed tomography (CT) scanned and radiographs were taken for more details. A preoperative radiograph showed approximately 3.0 mm of a separated instrument in the apical third of the distal root with a small portion of the file fragment extruded beyond the apex of the distal root with which a periapical lesion was associated (Figure 5a). There seemed to be a distal canal and mesiobuccal and mesiolingual canals that merged into one canal in the apical one third on the radiograph and CT images. All of the canals appeared to be under-filled and there seemed to be a tiny space between the file fragment and the tip of the filling materials in the distal root canal.

Figures 5a to 5k. Case 1: Utrasonic removal of a separated file extruded beyond the apex. 5a. Preoperative radiograph that shows a separated file in the apical third of the distal canal. 5b. File fragment was pushed more apically. 5c. File fragment back to the original place. 5d. Magnification of the file fragment in the apical one third of the canal. 5e. File fragment shifted to a coronal level. 5f. File fragment shifted to a more coronal level. 5g. File fragment shifted to the canal orifice. 5h. File fragment removal attempts completed. 5i. Retrieved file fragment measuring 3 mm. 5j. Postoperative radiograph showing ProRoot MTA (DENTSPLY Tulsa Dental Specialties) filled in the distal canal and gutta-percha filled in the mesial canals. 5k. Fiber post placed with composite resin.

Retreatment was commenced with the removal of the composite resin restoration. Gutta-percha filling material was removed with a BUC 1A ultrasonic tip (Obtura Spartan) and an EGPR (G. Hartzell & Son) hand instrument without solvent. A straight-line access to the fragment was obtained and the canal was enlarged with Nos. 2 and 3 GG burs so that the coronal portion of the separated file could be visualized through a dental operating microscope (ProErgo [Carl Zeiss Meditec]). Great caution against file extrusion was exercised with the use of ultrasonics. Therefore, the mesial canal wall in contact with the file fragment was carefully removed with the prototype ultrasonic tip as the distal canal was slightly curved in the mesial direction. However, the first ultrasonic attempt ended up pushing the file fragment further into the apical direction, as shown in Figure 5b. Vertical ultrasonic impact might have caused this, touching mainly on the severed surface of the file fragment. Then, the ultrasonic tip was inserted into the narrow space created between the file fragment and the mesial canal wall and was activated intermittently without touching the severed surface, so as to give ultrasonic vibration laterally to the file fragment. The first impact of ultrasonic activation on the side of the file fragment successfully shifted it back to the original place, as shown in Figures 5c and 5d. The file fragment was shifted to a more coronal level of the canal as ultrasonic activation focused on the side of the file fragment continued (Figures 5e and 5f). Intermittent ultrasonic vibration was given to the file fragment on the side until it was drawn completely inside the canal. Then, continuous ultrasonic activation was given laterally to the file fragment, and it came out of the canal to the orifice in 2 seconds (Figure 5g). A radiograph was taken to confirm retrieval of the file fragment (Figure 5h). The retrieved file fragment was 3 mm (Figure 5i). The removal procedure from the preparation to the completion of the separated file removal took approximately 4 minutes.
All the canals were negotiated with No. 10 K-files and all the canals were prepared with Greater Taper (DENTSPLY Tulsa Dental Specialties) rotary Ni-Ti instruments to the working length. The distal canal was filled with ProRoot MTA (DENTSPLY Tulsa Dental Specialties) as the diameter of the canal terminus was more than No. 70 (0.7 mm), and the other canals were obturated with 0.08 taper GT gutta-percha points (DENTSPLY Tulsa Dental Specialties) and Pulp Canal Sealer (SybronEndo), using a System B heat source (SybronEndo/Analytic) and Obtura (Obtura Spartan) for backfill (Figure 5j).
At the subsequent visit, fiber posts were placed in all of the canals with dual-cured composite resin, and a postoperative radiograph was taken (Figure 5k). The patient returned to the referring dentist for the final restoration.

CASE 2
A 36-year-old female, needing retreatment of the right mandibular second molar, was referred to our endodontic office. There was no significant medical history. She complained of a having sensitivity upon chewing for the past 3 months. Clinical examination revealed that the tooth was sensitive to percussion. The periodontal probing was within normal limits. Radiographic examination showed a separated file obviously longer than 4.5 mm in the coronal third of the mesial canal extending to a few millimeters beyond the apical foramen, and there was absence of obturation material in the apical one third of both the mesial and distal canals (Figure 6a). There was no evidence of periapical radiolucency in both the distal and the mesial roots.

Figures 6a to 6h. Case 2: Removal of a separated file with a loop device. 6a. Preoperative radiograph showing a long separated file in the apical one third of the mesial canal. 6b. Magnification of the separated file in the mesiobuccal canal upon gutta-percha removal. 6c. Magnification of the separated file on preparation for file removal. 6d. Radiograph after removal of the separated file. 6e. Retrieved file, measuring 6.5 mm. 6f. Postoperative radiograph showing the mesial canals filled with gutta-percha and the distal canal with ProRoot MTA. 6g. Postoperative radiograph showing fiber posts placed in mesial and distal canals with dual-cured composite resin.

Endodontic treatment was initiated by removing the metal crown and metal core material with a both a carbide bur and a high-speed diamond bur. Gutta-percha filling material was removed with a BUC 1A ultrasonic tip and EGPR hand instruments designed exclusively for gutta-percha removal. Upon gutta-percha removal, the separated file around the curve was already visible in the mesiobuccal canal under the microscope (as shown in Figure 6b), and the preparation for separated file removal was commenced using the prototype ultrasonic tip as described previously in the “instrument removal method” section. However, ultrasonic file removal attempts were not successful due to the long length of the separated file, even though it was observed being shaken loose from the canal (Figure 6c). Instead of continuing with the ultrasonic attempts to remove dentin apically for more than 5 minutes, the loop device was immediately used. First the loop was adjusted to the size of the file fragment, brought into the canal, placed over it, and fastened. The separated file was initially resistant to the pulling motion with the loop device, and it was finally removed by manipulating the device in all directions. A radiograph was taken for the confirmation of the separated file removal (Figure 6d). The removal procedure from the preparation to the completion of the separated file removal took approximately 7 minutes, and the retrieved file fragment measured 6.5 mm (Figure 6e). All the canals were prepared with Greater Taper rotary Ni-Ti instruments. The distal canal was filled with ProRoot MTA, as the diameter of the canal terminus was already more than No. 70 (0.7 mm), and the other canals were obturated with 0.10 taper GT gutta-percha points and Pulp Canal Sealer using a System B heat source and Obtura for backfill (Figure 6f).
At the subsequent visit, fiber posts were placed in all of the canals with dual-cured composite resin, and a postoperative radiograph was taken (Figure 6g). The patient returned to the referring dentist for the final restoration.

DISCUSSION
In spite of aggressive-looking removal attempts with ultrasonics, the amount of dentin removal in both of the cases seems to be minimal on the postoperative radiographs, and it is hardly distinct in comparisons with pre- and postoperative radiographs to tell which radiograph is preoperative. Ledge formation is generally obvious in curved canals on postoperative radiographs, which usually make it easy to tell what happened to the canals when it comes to separated file removal attempts. Several studies showed that ledges were inevitably created in the process of file removal attempts because of the staging platform.^7,16 However, there seems to be no distinct differences in canal ledge formation between preoperative and postoperative radiographs for both of the cases, where the separated files were located in the apical one third of curved canals. It seems as if dentin sacrifice for separated file removal had been a part of root canal preparation.
The results of these cases showed the instrument removal method performed on those 2 cases required as little dentin sacrifice as normal root canal preparation does. Dentin sacrifice for file removal attempts should be minimized, as it has been reported that canal enlargement of 40% to 50% of the root width increased susceptibility to vertical fracture.²³ It was possible that the use of greater taper files might have smoothed out the ledges when used to the canal terminus during the instrumentation procedure. Overall, the results obtained from the 2 cases showed preservation of the canal anatomy. Separated file removal time was 4 minutes for Case 1, and 7 minutes for Case 2. Suter et al²⁴ recommended that removal attempts of fractured instruments from root canals should not exceed 45 to 60 minutes because extended time may lead to iatrogenic errors such as root perforation and fracture. He suggested that the lowered success rate could result from operator fatigue or over-enlargement of the canal. It has been reported that the time needed to remove a separated instrument using the Masserann technique varied from 20 minutes to several hours, the time required from ultrasonic techniques varied from 3 to 40 minutes,16 while the method used for these case reports using the prototype ultrasonic tips and the loop device took only 4 to 7 minutes, which showed that the removal time here was consistent with other case reports using the same techniques.^21,22

CLOSING COMMENTS
Based on the complexity of the cases presented, this technique with the prototype tips and the loop device demonstrated high efficiency in the removal of separated files. For the past 2 years, I have removed 116 separated files from root canals with the same technique as described in this article. From the way I felt during the removal attempts, an adequate technique with as small instruments as possible may be a key to success.

References

1. Patiño PV, Biedma BM, Liébana CR, et al. The influence of a manual glide path on the separation rate of Ni-Ti rotary instruments. J Endod. 2005;31:114-116.
2. Ankrum MT, Hartwell GR, Truitt JE. K3 Endo, ProTaper, and ProFile systems: breakage and distortion in severely curved roots of molars. J Endod. 2004;30:234-237.
3. Nagai O, Tani N, Kayaba Y, et al. Ultrasonic removal of broken instruments in root canals. Int Endod J. 1986;19:298-304.
4. Glick DH, Frank AL. Removal of silver points and fractured posts by ultrasonics. J Prosthet Dent. 1986;55:212-215.
5. Krell KV, Fuller MW, Scott GL. The conservative retrieval of silver cones in difficult cases. J Endod. 1984;10:269-273.
6. Gettleman BH, Spriggs KA, ElDeeb ME, et al. Removal of canal obstructions with the Endo Extractor. J Endod. 1991;17:608-611.
7. Ward JR, Parashos P, Messer HH. Evaluation of an ultrasonic technique to remove fractured rotary nickel-titanium endodontic instruments from root canals: an experimental study. J Endod. 2003;29:756-763.
8. Hülsmann M, Schinkel I. Influence of several factors on the success or failure of removal of fractured instruments from the root canal. Endod Dent Traumatol. 1999;15:252-258.
9. Ruddle CJ. Nonsurgical retreatment. In: Cohen S, Burns RC, eds. Pathways of the Pulp. 8th ed. St. Louis, MO: Mosby; 2002:875-929.
10. Shen Y, Peng B, Cheung GS. Factors associated with the removal of fractured NiTi instruments from root canal systems. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2004;98:605-610.
11. Dominici JT, Clark S, Scheetz J, et al. Analysis of heat generation using ultrasonic vibration for post removal. J Endod. 2005; 31:301-303.
12. Gluskin AH, Ruddle CJ, Zinman EJ. Thermal injury through intraradicular heat transfer using ultrasonic devices: precautions and practical preventive strategies. J Am Dent Assoc. 2005;136:1286-1293.
13. Hashem AA. Ultrasonic vibration: temperature rise on external root surface during broken instrument removal. J Endod. 2007; 33:1070-1073.
14. Sweatman TL, Baumgartner JC, Sakaguchi RL. Radicular temperatures associated with thermoplasticized gutta-percha. J Endod. 2001:27:512-515.
15. Ruddle CJ. Nonsurgical retreatment. J Endod. 2004;30:827-845.
16. Souter NJ, Messer HH. Complications associated with fractured file removal using an ultrasonic technique. J Endod. 2005;31:450-452.
17. Madarati AA, Qualtrough AJ, Watts DC. Efficiency of a newly designed ultrasonic unit and tips in reducing temperature rise on root surface during the removal of fractured files. J Endod. 2009;35:896-899.
18. Marsicovetere ES, Clement DJ, del Rio CE. Morphometric video analysis of the engine-driven nickel-titanium Lightspeed instrument system. J Endod. 1996;22:231-235.
19. Terauchi Y. In vitro evaluation on secondary fracture of separated Ni-Ti files by ultrasonic vibration. Session of the Japan Endodontic Association; May 28, 2008; Chiba, Japan.
20. Terauchi Y. In vivo evaluation of the relationship between the lengths of intracanal separated instruments and the time to remove them with ultrasonics. Session of the Japan Endodontic Association; July 25, 2010; Tokyo, Japan.
21. Terauchi Y, O’Leary L, Suda H. Removal of separated files from root canals with a new file removal system: case reports. J Endod. 2006;32:789-797.
22. Terauchi Y, O’Leary L, Kikuchi I, et al. Evaluation of the efficiency of a new file removal system in comparison with two conventional systems. J Endod. 2007;33:585-588.
23. Wilcox LR, Roskelley C, Sutton T. The relationship of root canal enlargement to finger-spreader induced vertical root fracture. J Endod. 1997;23:533-534.
24. Suter B, Lussi A, Sequeira P. Probability of removing fractured instruments from root canals. Int Endod J. 2005;38:112-123.

Dr. Terauchi is one of the few endodontists practicing in Japan at the CT and Microendodontic Center. He is well known for his expertise in removing fractured endodontic files from root canals. He received the Presidential Award from the Japan Endodontic Association in both 2001 and 2008. He is a part-time lecturer at Tokyo Medical and Dental University. He can be reached via e-mail at yterauchi@intellident.or.jp.

Disclosure: Dr. Terauchi reports no disclosures.