Clin Oral Invest (2005) 9: 84 – 90 90 DOI 10.1007/s00784-004-0299-7
ORIGINAL ARTICLE
A. Sahafi . A. Peutzfeldt . G. Ravnholt . E. Asmussen . K. Gotfredsen
Resistance to cyclic loading of teeth restored with posts
Received: 6 May 2004 / Accepted: 8 November 2004 / Published online: 4 March 2005 # Springer-V Springer-Verlag erlag 2005 Abstract Thi Thiss in vitr vitroo stu study dy eva evalua luated tedthe the eff effect ect of pre presen sence ce
Keywords Adhesion . Dowel . Intermittent loading .
crowned showed significantly higher cyclic loading than teeth with cores or with postresistance and cores.toNo significant differences were found between teeth restored with cores only or with post and cores, irrespective of surfacetreatment of the posts. Teeth restored with parallel-sided cast post (ParaPost XP) and cores showed significantly higher resistance to cyclic loading than teeth with either tapered cast posts or untreated prefabricated posts of titanium alloy (ParaPost XH) or glass fiber composite (ParaPost Fiber White). No significant difference was found between teeth restored with parallel-si parallel-sided ded cast post and cores and teeth restored with untreated prefabricated posts of zirconia (Cerapost). Surface treatment of posts significantly increased the resistance to cyclic loading compared with untreated posts. When posts are used, surface treatment is recommended.
bility for final restorations restor ations endodontical ntically treated treate teeth [19 19,, 31 31]. ]. Ge Gene nera rall llyy, tw two o ty type pesof s ofendodo post po st an and d co core relysy syst stem emsds ex exis ist: t: custom-made post and cores and prefabricated posts with resin composite cores. Individually cast post and cores are normally cast from metal alloys. As for prefabricated posts, thesee are eit thes either her met metall allic ic pos posts ts suc suchh as sta stainl inless ess ste steel, el, tit titani anium um alloy and noble metal posts, which traditionally have been lutedd wit lute withh zin zincc phos phosphat phatee cem cement ent,, or non non-me -metal tallic lic post postss such as posts of zirconia zirconia and carbon fiber or glass fiber fiber reinforced reinforced resinn com resi compos posite ite,, whi which ch are int intend ended ed to be adh adhesi esivel velyy bon bonded ded in the root canal. The effect of type of post and core on resistance to loadingg of re in rest stor ored ed te teet ethh ha hass be been en in inve vest stig igat ated ed in se seve vera rall in vi vitr troo studies [1 [1, 5 5,, 12 12,, 14 14,, 21 21]] and conflicting results have been reported. Isidor et al. [14 [ 14]] investigated the resistance to cyclic loading of teeth restored with either individually cast or prefabricated posts and found higher resistance to cyclic loading of the teeth restored with prefabricated posts. In another in vitro study, Butz et al. [5 [ 5] compared the resistance to cyclic and static loading of teeth restored with either individually cast or prefabricated posts and found no differences in resistance to loading of the restored teeth. Differences in resistance to cyclic loading have been shown for various types of prefabricated posts: the use of carbon, quartz, or carbon-quartz fiber posts has been found to result in lower failure rates than posts of zirconia or metal, most probably due to the bonding of the posts to the root canal [15 [15,, 20 20]. ]. As regards the shape of posts, two fundamental shapes exist: tapered and parallel. A number of in vitro and in vivo studies have shown superior retention, higher resistance to
of post, presence of core, and of shape, type, and surface Posts . Resistance to fracture . Weakened teeth treatm tre atment ent of of pos posts ts on resi resista stance nce to cycl cyclic ic load loading ing of of crowne crownedd human teeth. For all teeth, crowns designed without ferrule were cast in sterling silver and luted with resin cement Introduction (Pan (P anav avia ia F) F).. Ea Each ch to toot othh un unde derw rwen entt cy cycl clic ic lo load adin ingg of 60 6000 N at two loads per second until failure. Teeth that had only been Post and cores are often used to provide retention and sta-
A. Sahafi (*) . K. Gotfredsen Department of Prosthetic Dentistry, School of Dentistry, University of Copenhagen, Nörre Allé 20, DK-2200 DK-22 00 Copen Copenhagen hagen N, Denmark Denmark e-mail:
[email protected] Tel.: +45-35326745 Fax: +45-35326505 A. Sahafi . A. Peutzfeldt . E. Asmussen Department of Dental Materials, School of Dentistry, University of Copenhagen, Copenhagen, Denmark G. Ravnholt Department of Prosthetic Dentistry, School of Dentistry, University of Aarhus, Aarhus, Denmark
cyclic cyc lic loa loadin ding, and hig higher her suc succes ratee of rat teeth th res restor tored ed with wi th parallel-side parallel-sided dg,posts compare compared dcess to steeth restee restored tored with tapered tapered posts [7 [7, 14 14,, 16 16,, 25 25,, 34 34,, 36 36,, 42 42]. ].
85
The fracture resistance of endodontically treated teeth is posts. Consequ Consequently ently,, the aim of this in vitro study was to largely dependent on the amount of remaining tooth struc- evaluat evaluatee the effect of posts, of core, and of shape, type, and ture [18 [18,, 43 43]. ]. Preparation of the root canal reduces remain- surface treatment of posts on the resistance to cyclic loading ing tooth structure and the use of a post may weaken the of crowned human teeth. tooth and increase the risk of root fracture by uneven stress distribution along the length of post. A number of in vitro studies have investigated the fracture strength of teeth re- Material and methods stored with and without post and cores: either no effect or a positive effect on fracture strength of teeth was found as a One hundred extracted, healthy human maxillary incisors consequence of omitting a post [3 [3, 12 12,, 43 43]. ]. and canines and mandibular canines with a root length of In attempts to maximize the retention of posts and the fracture strength of roots, luting of posts with resin cement as wel welll as sur surfac facee treatm treatment entss of pos posts ts hav havee bee beenn stu studie diedd [10 [10,, 17,, 22 17 22,, 23 23,, 26 26,, 27 27,, 32 – 34, 34, 41 41,, 44 – 46]. 46]. One surface treatment, the tribochemical silicate-coating silicate-coating system CoJet (3 M ESPE), has been found to create an effective bonding of resin cement to various types of prefabricated posts [32 [ 32]. ]. The system uses silicate-coated alumina particles to sand blast the surface prior to application of silane and resin cement. Sandblasting with the silicate-coated alumina particles produces a high spot heat, which together with the blasting pressure pressure results results in welding welding of the silicate silicate layer onto the surface [9 [9] and subsequent silanization enhances the bond strength of resin cement to the treated surface [6]. These studies have investigated the fracture strength of
at least 10 mm were kept in an antimicrobial preservative (0.5% Chloramine T) after extraction. The root of each tooth was roughened by green grit (SiC) No. 48 (Hallvard Foss & Co., Fetsund, Norway) and following application of an adhesive system Optibond Solo Plus (Kerr, Orange, CA, USA), a ball of resin composite Flow line (Heraeus Kulzer, Inc., NY, USA) was applied to the apex and light cured for 30 s. The surface of the root and composite ball was roughened by green grit and coated with a layer of silicone rubber RTV 11 GE Silicones (GE Bayer Silicones, Bergen op zoom, The Netherlands). After drip drying and curing under humid conditions for 1 d, this artificial periodontal ligament had a thickness of approximately 60 μ m [13 13]. ]. The teeth were mounted in a cylindrically prepared cavity in a prefabricated acrylic block (d=10 mm, l=15 mm)
teeth restored posts, the bond strength of to po post sts, s, or th theewith reten re tentiv tivee st stre rengt ngth h of po postslute stsluted d inresin roottcement roo canals can als.. In an attempt to simulate the complex, clinical conditions more closely, in vitro studies of resistance to cyclic loading of te teet ethh re rest stor ored ed wi with th po post st an andd co core ress wi with th th thee us usee of co cove veri ring ng crowns cro wns have have been carried carried out [13 – 15]. 15]. How Howeve ever, r, these these latt latter er studies have not investigated the effect of surface treatment of posts. The pre presen sentt stu study dy tes tested tedthe the fol follow lowing inghyp hypoth otheses eses:: 1) teet teethh restored with only crowns have higher resistance to cyclic loadingg than teeth also restored with cores or with post and loadin cores; cor es; 2) tee teeth th res restor tored ed wit withh only onlycor cores es hav havee hig higher her res resist istanc ancee to cy cycl clic ic lo load adin ingg th than an te teet ethh al also so re rest stor ored ed wi with th pos postt an andd co core res; s; 3) teeth restored with parallel-sided posts have higher resistanc ta ncee to cy cycl clic ic lo load adin ingg th than an te teet ethh re resto store redd wi with th ta tape pered red po post sts; s; 4) teeth restored with surface-treated posts have higher resistance to cyclic loading than teeth restored with untreated posts;; 5) teeth restored with posts with prefabricated prefabricated posts have higher higher resistance to cyclic loading than teeth restored with cast
and fixed with auto-curing acrylic Kulzer, Inc., NY, USA), which wasresin left toPaladur cure for(Heraeus 60 min. The ball of resin composite helped secure retention of the tooth in the acrylic block. The teeth were randomly distributed into ten experimental groups, each consisting of ten teeth (Table 1 (Table 1). ). Teeth planned to receive a crown (without post and core) weree wet wer wet-se -secti ctione onedd hor horizo izonta ntally lly in the cor corona onall par partt and the thenn prepared. The height of the preparation was 3 mm, the angle of convergence was approximately 15°, and the height of root outside the acrylic block to the crown margin was 5 mm. Teeth planned to receive a core but no post were wetsectioned to leave 5 mm of root outside the acrylic block. Coress of aut Core auto-cu o-curin ringg resi resinn com composi posite te Cle Clearfi arfill cor coree (Ku (Kurara rarayy, Osaka, Japan) were then bonded by the use of an adhesive system Clearfil LinerBond 2 V (Kuraray, Osaka, Japan). After a setting time of 20 min, the cores were prepared to receive a crown. The heights of prepared cores were 3 mm,
Table 1 Experimental groups (n =10)
Type ype of post post an andd core core rest restor orat atio ionn
Type, ype, ma mate teri rial al,, and and shap shapee of po post st
Surf Surfac acee trea treatm tmen entt of post post
Type ype of core core
No post, no core No post Tapered post ParaPost XP ParaPost XH ParaPost XH ParaPost Fiber White ParaPost Fiber White Cerapost Cerapost
None None Cast, sterling silver, tapered Cast, sterling silver, parallel-sided Prefabricated, titanium alloy, parallel-sided Prefabricated, titanium alloy, parallel-sided Prefabricated, glass fiber, parallel-sided Prefabricated, glass fiber, parallel-sided Prefabricated, zirconia, parallel-sided part Prefabricated, zirconia, parallel-sided part
None None None None N Noone Cojet system None Cojet system None Cojet system
None Resin composite Cast in sterling silver Cast in sterling silver Resin composite Resin composite Resin composite Resin composite Resin composite Resin composite
86
Post XP (Col PostXP (Coltèn tène/W e/Whale haleden dent,t, Mah Mahwah wah,, NJ, NJ,USA USA)) and tap tapered ered posts) post s) wer weree stu studie died. d. The compositi composition on and manufactur manufacturers ers of the investigated prefabricated posts and the cement are listed in Table 2 Table 2.. Teeth assigned to receive ParaPost XP, ParaPost XH, ParaPost Fiber White, or Cerapost (Ceraposts were used “upside-down ” in order for the Cerapost to be parallelsided as are ParaPost XH and ParaPost Fiber White) were prepared with the calibrated ParaPost ParaPost drill system to a final diameter of 1.4 mm. Te Teeth eth planned to receive a cast tapered
Fig. Fi g. 1 Schem Schematic atic drawing of root restor restored ed with prefa prefabricat bricated, ed, parallel-sided post and resin composite core. A = 5 mm; length of root protruding from acrylic block; B = 3 mm; height of resin composite core; C = 6 mm; length of post luted in the root canal. D = ball of resin composite. E = artificial periodontal membrane of
silicone
the angle of convergence was 15°, and the height of the ferrule (i.e. the vertical dentinal overlap of the crown) was 0 mm, i.e. the crowns were without ferrules. ferrules. Teeth planned to receive a post and core were wet-sectioned to leave 5 mm of root outside the acrylic block and restored with post and cores as described below and shown in Fig. 1 Fig. 1.. Three types of prefabricated posts (a titanium alloy post ParaPos Par aPostt XH (Col (Coltèn tène/W e/Whal halede edent, nt, Mah Mahwah wah,, NJ, USA USA), ), a glass fiber reinforced resin composite post ParaPost Fiber White (Coltène/Whaledent, Mahwah, NJ, USA), and a zirconia post Cerapost (Gebr. Brasseler, Lemgo, Germany)) and two shapes of cast post and cores (parallel-sided posts ParaTable 2 List of investigated investigated Post prefabricatedposts and cement
ParaPost XH
post wer weree prep prepared aredby by a tap tapered ereddril drilll Ger Gerlach lach(Geb (Gebrr. Bras Brassel seler er,, Lemgo,, Germ Lemgo Germany) any) to a fina finall coro coronal nal diameter diameter of 1.5 mm. The length of the prepared root canal was 7 mm. After preparation, the the canals were were rinsed with deionized water water for 2 min. For fabrication of the cast post and cores, polyether im pressions pressi ons in Impreg Impregum um Penta (3 M ESPE, Seefel Seefeld, d, German Germany) y) were made following post preparation. For teeth planned to receive ParaPost XP posts, the corresponding impression plasticc posts were placed in the post prepar plasti preparations ations,, and for teeth planned to receive tapered posts, loose-fitting plastic pins were used to stabil stabilize ize the impres impression sion mater material ial in the post preparations prepar ations.. The impres impressions sions were poured with Vel-Mi el-Mixx stone, ISO type IV (Kerr, Orange, CA, USA). Following the use of either ParaPost XP plastic burnout posts or waxed plasti plastic c pins, cores withwere the height and the convergence of 15° waxedofon3 mm the dies andangle cast of in sterling silver. The effect of the surface treatment CoJet (3 M ESPE, Seefeld, Germany) was tested by the use of prefabricated posts. CoJet treatment consisted consisted in air abrasion with an intraoral sandblasting device Dento-prep (Rønvig, Daugaard, Denmark) at 4 bar for 15 s using 30-μ m silicate-coated particles parti cles follo followed wed by silan silanee coati coating ng with ESPE-s ESPE-sil il (3 M ESPE, Seefeld, Germany) according to the manufacturer ’s instructions. Prio Pr iorr to lu luti ting ng of th thee po post sts, s, th thee ro root ot ca cana nals ls we were re dr drie iedd wi with th paper points Top Dent No. 45 (Svenska Dental Instrument, Instrument, Upplands Väsby, Sweden) and treated with ED primer (Kuraray, Osaka, Japan) according to the manufacturer ’s
Compositi ition according to manufacturer 90% tit titanium, 6% aluminu inum, 4% vanadium
ParaPost Fiber 42% glass fiber, 29% resin, 29% filler White Cerapost 94.9% ZrO2, 5.1% Y2O3 Cement Panavi Pan aviaa F
ED prime primerr
Manufacturer
Coltène/Whaledent, USA Coltène/Whaledent, USA Gebr. Brasseler, Germany
Kuraray, Japan Sil Silana anated ted bar barium ium gla glass, ss, sila silanat nated ed silic silica, a, sodi sodium um fluo fluoride ride,, BPO, BPO, photosensitizer, 10-methacryloyloxydecyl dihydrogen phosphate (MDP), dimethacrylate, hydrophobic and hydrophilic dimethacrylate, Bis-phenol A polyethoxy dimethacrylate 10-methacry 10-methacryloyloxy loyloxydecyl decyl dihydrogen dihydrogen phosph phosphate ate (MDP), (MDP), HEMA, Kuraray, Japan N-methacryl 5-aminosalicylic acid, sodium benzene sulfinate, N, N-diethanol p-toluidine, water
87
directions. The cement Panavia F (Kuraray, Osaka, Japan) was mixed according to the manufacturer ’s recommended procedure and applied in the root canal, and the post was luted to a length of 6 mm. The dual-curing Panavia F was light-cured for 20 s with a conventional quartz tungsten halogen curing unit XL 3000 (3 M ESPE, Seefeld, Germany). All specimens were allowed to set for 15 min. In the case of roots restored with prefabricated posts and resin com positee cores, posit cores,these these were produce producedd and prepar prepared ed as previo previously usly described. Crowns of sterling silver, which had occlusal surfaces with a 45° angle to the long axis of the teeth and a maximum height of approximately 7.5 mm, were cast. In the case of a crown without without post and core, crowns were luted with Panavia F following application of ED Primer. In the case ca se of a co core re on only ly (n (noo po post st)) or a po postand stand co core re,, cr crow owns ns we were re luted lu ted wi with th Pa Panav navia ia F wi with thou outt ap appl plic icat atio ionn of ED Pr Prim imer er.. Th Thee restored teeth were stored in water at 37°C for at least 1 d. The resistance to cyclic loading of the restored teeth was tested at 37°C using two loads of 600 N/s until failure. The teeth were kept humid during testing by means of wet cotton cloth cloth covering the the exposed roots. roots. The force was generatedbyanaircylinderatapressureof5.25bar.Theloading was perpendicular to the occlusal surface of the crown (i.e. 45°° an 45 angl glee to th thee lo long ng ax axis is of th thee to toot oth) h) as de desc scri ribe bedd by Is Isid idor or et al.the [13]. [13 ]. The test recorded the numberwhen of loads, and testing wasmachine automatically discontinued the system failed by loss of retention or fracture fracture of the root and/ or the post. Because of the non-normal distribution of the results, the median values were calculated for all experimental groups, and Mann-Whitney U-tests were used to analyze the results with corrections for multiple comparisons by the Bonferroni method [2 [2]. The level of significance was set to α =0.05. =0.05. Results
The results of the cyclic loading tests are presented in Table 3 Table 3.. Teeth restored with a crown without post and core gave significantly higher resistance to cyclic loading than teeth restored with cores only (no post) or with post and cores. There were no significant differences in resistance to cyclic loading between teeth restored with cores only (no post) and teeth restored with post and cores, irrespective of surface treatment of posts. Post shape had a significant effect in that teeth restored withh par wit parall allelel-sid sided ed cas castt pos postt and cor cores es (Par (ParaPo aPost st XP) sho showed wed signif sig nifican icantly tlyhigh higher er res resist istanc ancee to cycl cyclic ic load loading ing tha thann did didteet teethh rest re stor ored ed wi with th ta tape pere redd ca cast st po post st an andd co core res. s. Tee eeth th re resto store redd wi with th parallel-sid parall el-sided ed cast post and cores were also signif significantl icantlyy more resistant to cyclic loading than teeth restored with untrea unt reate tedd pre prefab fabric ricate atedd Para ParaPos Postt XH or Par ParaPo aPost st Fib Fiber er Whi White te posts and resin comp composite osite cores cores.. Howe However ver,, no signi significan ficant t difference was found between teeth restored with parallel-
Table 3 Media Mediann and range of load load units of 2× 600 N/sunti N/suntill failure
of the experimental groups (n =10)
Type of post restoration
Median
Range
No post, no core No post Cast tapered post Cast parallel-sided post (ParaPost XP) ParaPost XH, untreated
1.009.525 367.922 140.280 418.933
689.771 – 1.726.399 1.726.399 6.752 – 690.514 690.514 1.670 – 346.725 346.725 114.760 – 1.180.971 1.180.971
Pa Para raPo Post st XH XH,, Coje Cojett-tr trea eate tedd ParaPost Fiber White, untreated ParaPost Fiber White, Cojet-treated Cerapost, untreated Cerapost, Coje jett-ttreate tedd
190.524
1.517 – 708.662 708.662
542. 542.70 7077 169.837
29 298. 8.54 5400 – 1.033.161 845 – 497.218 497.218
322.948
8.946 – 758.216 758.216
435.520 495.806
485 – 689.260 689.260 178.342 – 797.674 797.674
No differences were found in resistance to cyclic loading between the three types of prefabricated posts. This held true when the posts remained untreated as well as when the posts had been surface-treated. Consequently, all results obtained with untreated and with surface-treated posts, respectiv respectively ely,, were pooled. pooled. Statistical Statistical analysis analysis showed thatcyclic surface surfa ce treatm treatment significantl signif icantlyywith increased increa sed the resist resistance ance to loading ofent teeth restored prefabricated posts. The vast majority of the restored teeth failed due to root fracture (94%). Fracture of post was observed for 3% of the restored teeth while loss of retention of post or core was responsible for the remaining 3% of the failures. Discussion
This study found that teeth with a moderate reduction of tooth structure, i.e. teeth restored with crowns without post and cores, had higher resistance to cyclic loading than had teeth with a pronounced reduction of tooth structure, i.e. teeth restored with either a core or a post and core, and that the use of a post did not improve the resistance to cyclic loading. It was also found that the resistance to cyclic loading of teeth teeth restored with with post and cores was influenced by the shape, type (cast versus prefabric prefabricated), ated), and surface treatment of the post. Regarding the influence of the amount of remaining tooth structu stru cture, re, the dif differ ference encefoun foundd bet betwee weenn teet teethh wit withh a mod moderat eratee reduction of tooth structure and teeth with a pronounced redu re duct ctio ionn of to toot othh st stru ruct ctur ure, e, is in ag agre reem emen entt wi with th a nu numb mber er of other studies [8 [8, 29 29,, 36 – 38, 38, 43 43]. ]. Post and cores are usually inserted before the final restoration of destructed, endodontica ti call llyy tr trea eate tedd te teet eth. h. The fu func ncti tion on of th thee po post st in th thee roo roott cana canall is to improve the retention of the core. However, the preparation of the root canal for the post weakens the root and thereby increases the risk of root fracture. In vitro and in vivo studies of fracture strength and failure mode of teeth
sided cas sided castt pos postt and andcor cores es and tee teeth th res restor tored ed wit withh pre prefab fabric ricate atedd restored with various post systems (individually cast post Cerapost posts. and cores and prefabricated metallic or zirconia posts with resin composites cores) showed that the use of posts not
88
only decreased the fracture strength or success rate of the teeth as compared to no post but also resulted in unrestorable fracture of the teeth [3 [3, 12 12,, 37 37,, 43 43]. ]. In the present study, in which the crowns had no ferrule, no differences were found in resistance to cyclic loading between teeth restor st ored ed wi with th a co core re on only ly (n (noo po post st)) an andd te teet ethh re rest stor ored ed wi with thaa po post st and core. The explanation for this finding may be an effective adhesion between the resin composite cores and dentin combined with no weakening effect of root canal preparation. Signific Sign ificantl antlyy high higher er resi resistan stance ce to cycl cyclic ic load loading ing was found for teeth restored with parallel-sided cast post and cores cor es tha thann for fortee teeth th res restor tored ed wi with th tap tapere eredd cas castt pos postt andcores andcores.. This finding supports the in vitro cyclic loading study of Isidor et al. [15 [15], ], who found teeth restored with prefabricated parallel-sided post and cores to have higher resistance to cyclic loading than teeth restored with tapered cast post and cores. The authors suggested that the higher resistance of teeth restored with parallel-sided posts is due to a more even stress distribution of parallel-sided versus tapered post andcores and cores.. The even evenstre stress ss dist distribu ribution tionof of para parallel llel-sid -sided ed posts has also been confirmed by a number of photoelastic stress analyses anal yses[[4, 11 11,, 39 39]. ]. Th Thee fa fact ct th that at hig higher her res resist istanc ancee to cyc cyclic lic loading was obtained when parallel-sided posts were used is also supported by two in vivo studies in which the
modulus of elasticity: ParaPost XH = 112 GPa, ParaPost Fiber White = 29.2 GPa, and Cerapost = 200 GPa. A regression analysis performed on number of cyclic loadings until failure for the three posts ( n =30) showed a significant influence of the modulus of elasticity of the prefabricated posts ( p
