Outcomes of root canal treatment in Dental Practice-Based ...


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Outcomes of root canal treatment in Dental Practice-Based ...

  1. 1. EndodonticsOutcomes of root canal treatment in DentalPractice-Based Research Network practicesGregg H. Gilbert, DDS, MBA   n  Ken R. Tilashalski, DMD Mark S. Litaker, PhD Sandre F. McNeal, MPH   n    n Michael J. Boykin, DMD, MS   n  Allen W. Kessler, DMD, for the DPBRN Collaborative GroupThis study sought to quantify the incidence of root canal treatment a significant predictor of treatment failure (in other words, patients(RCT) failure and identify its predictors in root canals that were who did not receive a permanent restoration were more likely toperformed or referred by general dentistry practices in a practice- experience RCT failure), whether failure was determined clinicallybased research network (PBRN). This retrospective cohort study or radiographically. This study of PBRN practices suggests a higherinvolved 174 endodontically treated teeth. Mean duration from initial failure rate compared with studies that utilized highly controlledtherapy to follow-up was 8.6 years. Permanent restorations were environments or populations with high levels of dental insurance.ultimately placed in 89% of teeth, although 18% of teeth were Received: February 4, 2009ultimately extracted anyway. Receiving a permanent restoration was Accepted: May 5, 2009E stimates of the success of non- how the definition of treatment fail- a shorter amount of time in these surgical root canal treatment ure affects interpreting the results. non-academic settings, compared (RCT) have varied widely. Practice-based research networks to conventional clinical researchMany studies and endodontic (PBRNs) offer a venue for con- performed in academic settings bytextbooks have reported very ducting RCT outcome studies in academic dentists.7-9 The Dentalhigh success rates; however, RCT community private practice settings. Practice-Based Research Networkoutcome studies usually have been PBRNs have existed in the U.S. since (DPBRN) originated in Alabamaconducted in academic settings or the 1970s.6 These networks bring and expanded to include four otheramong patient populations with dentists and academic researchers regions.12 The Alabama portion ofhigh levels of dental insurance together to develop and answer the DPBRN served as the setting forcoverage.1-5 As a result, it may not relevant research questions that the present study.be possible to apply these results to can directly impact routine clinical This study sought to quantify thecommunity-based private practices. practice. These networks offer unique incidence of RCT failure in privateRCT outcome studies are warranted advantages when conducting research practice settings, using three differ-for these community settings. about routine clinical care and ent definitions of treatment failure; In addition, these studies have when investigating how to improve to quantify the typical length ofshown wide variations in terms of the quality of clinical care, while time required for an endodonticallyhow treatment failure is defined. fostering the sharing of information treated tooth to receive a permanentSome studies have relied entirely between practitioners.7-10 restoration in a private practiceon information gleaned from Traditionally, clinical research has setting; to test the hypothesis thatradiographs. Others have excluded been conducted in academic set- receipt of a permanent restoration isextracted or fractured teeth, using tings; however, less than 1% of the a significant predictor of treatmentthe justification that extractions, U.S. population receives its health failure; and to test the hypothesisfractures, and restoration failures care in that type of setting.11 Studies that other factors are significantwere not directly caused by the that are initiated and developed predictors of treatment failure.RCT and therefore should not be by non-academic practitioners and A retrospective cohort study wasincluded as part of the assessment.5 conducted in their offices (as is the designed to meet these objectives.It could be instructive to conduct case with PBRNs) should produceRCT outcome studies within actual results that are more relevant to Materials and methodscommunity settings using multiple these clinicians and should lead Practices in the community weredefinitions of failure, to better assess to improved clinical treatment in identified based on the Alabama28 January/February 2010 General Dentistry www.agd.org
  2. 2. portion of the DPBRN. Dentist of the offices, screened the charts, with inadequate resolution or filmspractitioner-investigators were abstracted and entered data, and whose imaging of periapical struc-recruited through mailings to scanned radiographs. tures was insufficient.licensed dentists. As part of Dental charts were used to The two evaluators were cali-enrollment in the DPBRN, all abstract dates of all visits in which brated on the use of the periapicalpractitioner-investigators complete any treatment was administered index (PAI) (1–5 scale) using ana 101-item enrollment question- to endodontically treated teeth. authorized reference set providednaire about themselves and their RCT was defined as ADA codes by Orstavik.15 Evaluation of thepractice characteristics. Surveys in 3310, 3320, or 3330; in addition, study radiographs was done inde-the literature have demonstrated the researchers recorded the tooth pendently, although the evaluatorsthat DPBRN dentists have much in or areas treated; ADA codes for re-evaluated any disagreementscommon with dentists at large.13 all procedures performed on that jointly and reached a consensus. tooth; a description of the proce- All radiographs were digitizedRecruitment methods dure code (to ensure that the code before measurements were taken.Practices consisting of general matched its description); whether From these images, the evaluatorsdentists were identified for inclu- the tooth served as an abutment recorded the number of pre- andsion in the current study if they for a removable partial denture post-RCT proximal contacts (eitherwere enrolled in the DPBRN, had (RPD), overdenture, or fixed partial 0, 1, or 2) and pre- and post-RCTa practice address in Alabama, and denture; whether the RCT was measurements of the PAI.performed or referred more than done by going through an existing Pre- and post-RCT bone loss were20 root canals each month (based prosthetic crown; whether the RCT measured by using a digital subtrac-on a question posed in the enroll- was performed by a general dentist tion laboratory. A measurement wasment questionnaire). From the 25 or an endodontist; the date of data made from the crestal bone heightpractitioner-investigators who met entry; and the patient’s demo- to the root apex, which served as thethese criteria and expressed a desire graphic characteristics.14 Data were numerator. The denominator wasto be included in the study, 13 gathered from a total of 174 teeth in calculated by measuring the distancewere included in the study after a 84 subjects. from the cemento-enamel junctionvisit to the practice confirmed that to the root apex.the practice’s dental charts would Measurements made Using the pre-RCT radiographconsistently include the treatment from radiographs only, the evaluators recordedinformation and radiographs Two endodontic faculty evaluators whether there was radiographicnecessary for the study. A follow-up (KT and MB) made measurements evidence of caries at access. Usingdate was arranged for the research from radiographs taken before the post-RCT radiographs only,assistant to screen charts for poten- and after each RCT procedure. the evaluators recorded whethertial cases. Because most practices No identifying information was there was any radiographic evidencehad hundreds of charts, a random provided with these radiographs so of recurrent caries adjacent to thenumber generator was used to iden- as to mask any characteristics of the restoration that restored the RCT;tify a random start for chart review. subjects. All but one of the preop- whether the RCT filling was inad- erative radiographs were periapi- equate; the presence of a permanentChart abstraction procedures cal; the other was a panoramic restoration; and if the restorationA research assistant whose back- radiograph. Of the postoperative was inadequate.ground included training as a films, 78% were panoramic andcertified dental assistant was trained 22% were periapical. Each evalu- Definitions of failure andin chart abstraction, using dental ator made a determination as to follow-up periodcharts from dental schools and whether the resolution of the radio- One measure of failure was whetherprivate practices. Following several graph was sufficient to make an the endodontically treated tooth wastraining sessions, she achieved stan- accurate measurement. Of the 174 ultimately extracted. Such occur-dardization in proper abstraction endodontically treated teeth, 115 rences were documented, as were theprocedures specific to this study and had radiographic measurements number of days between the date thatachieved high intra-rater reliability. made. The excluded radiographs RCT was completed and the dateThis research assistant contacted all were either panoramic radiographs when the extraction was performed. www.agd.org General Dentistry January/February 2010 29
  3. 3. Endodontics  Outcomes of root canal treatment in DPBRN practices Table 1. Number (and percentage) of endodontically treated teeth that failed, according to definition of failure and the tooth’s characteristic or clinical circumstance (n = endodontically treated teeth; n1 = radiographically measured teeth). Teeth that were Teeth that experienced Teeth that displayed evidence Characteristics extracted (%) clinical failure (%) of radiographic failure (%) Overall ( n = 174; n1 = 115) 31 (18) 35 (20) 30 (26) Gender of RCT patient Female ( n = 86; n1 = 53) 16 (19)* 20 (23)* 15 (28)* Male ( n = 88; n1 = 62) 15 (17) 15 (17) 20 (32) Race of RCT patient Non-white ( n = 13; n1 = 10) 4 (31)* 4 (31)* 4 (40)* White ( n = 161; n1 = 105) 27 (17) 31 (19) 31 (30) Tooth type Anterior ( n = 30; n1 = 19) 2 (7) † 3 (10) † 3 (16) † Premolar ( n = 57; n1 = 36) 11 (19) 12 (21) 11 (31) Molar ( n = 87; n1 = 60) 18 (21) 20 (23) 21 (35) Tooth received a permanent restoration during follow-up No ( n = 20; n1 = 9) 8 (40) † 9 (45) ‡ 5 (56)* Yes ( n = 154; n1 = 106) 23 (15) 26 (17) 30 (28) Tooth was an RPD abutment at the time RCT was initiated No ( n = 169; n1 = 112) 28 (17) † 32 (19)* 33 (29)* Yes ( n = 5; n1 = 3) 3 (60) 3 (60) 3 (67) Tooth was a bridge abutment at the time RCT was initiated No ( n = 163; n1 = 109) 30 (18)* 34 (21)* 34 (31)* Yes ( n = 11; n1 = 6) 1 (9) 1 (9) 1 (17) RCT was performed by accessing through an existing crown No ( n = 97; n1 = 65) 18 (19)* 21 (22)* 20 (31)* Yes ( n = 77; n1 = 50) 13 (17) 14 (18) 15 (30) Number of proximal contacts at the time RCT was initiated None ( n = 0; n1 = 5) 0 (0) ‡ 0 (0) ‡ 4 (80) One or two ( n = 0; n1 = 110) 19 (17) 21 (19) 31 (28) A second measure of failure and either the date when clinical 3, 4, or 5 based on the post-RCTknown as clinical failure was failure occurred or, when clinical radiograph. The radiographicapplied to endodontically treated failure did not occur, the date of survival time was defined as theteeth that were extracted or the last clinical visit. number of days between the com-received endodontic retreatment A tooth was classified as a pletion of RCT and either the date(ADA codes 3346, 3347, 3348, or radiographic failure if the endo- that radiographic failure occurred3410). The clinical survival time dontically treated tooth had been or, when radiographic failure didwas defined as the number of days extracted, received endodontic not occur, the date of the tooth’sbetween the completion of RCT retreatment, or had a PAI score of last radiograph.30 January/February 2010 General Dentistry www.agd.org
  4. 4. Teeth that were Teeth that experienced Teeth that displayed evidence Characteristics extracted (%) clinical failure (%) of radiographic failure (%) Tooth had radiographic evidence of caries at the time of RCT access No ( n = 0; n1 = 97) 16 (16)* 18 (19)* 29 (30)* Yes ( n = 0; n1 = 18) 3 (17) 3 (17) 6 (33) Periapical index score at the time RCT was initiated 1 ( n = 0; n1 = 26) 3 (12)* 4 (15)* 7 (27)* 2 ( n = 0; n1 = 33) 2 (6) 3 (9) 6 (18) 3 ( n = 0; n1 = 37) 10 (27) 10 (27) 14 (38) 4 ( n = 0; n1 = 17) 3 (18) 3 (18) 7 (41) 5 ( n = 0; n1 = 2) 1 (50) 1 (50) 1 (50) Recurrent decay was present adjacent to the restoration on the post-RCT radiograph No ( n = 0; n1 = 99) 9 (9) ‡ 11 (11) ‡ 24 (22) ‡ Yes ( n = 0; n1 = 7) 2 (29) 2 (29) 3 (43) Missing ( n1 = 9) § The endodontic fill was judged to be poor based on the post-RCT radiograph No ( n = 0; n1 = 90) 9 (10) ‡ 10 (11) ‡ 19 (21) † Yes ( n = 0; n1 = 16) 2 (13) 3 (19) 8 (50) Missing ( n1 = 9) § Restoration had an overhang or open margin on the post-RCT radiograph No ( n = 0; n1 = 104) 9 (9) ‡ 11 (11) ‡ 25 (24) ‡ Yes ( n = 0; n1 = 2) 2 (100) 2 (100) 2 (100) Missing ( n1 = 9) § * Not statistically significant ‡ Statistically significant at p < 0.05 ‡ Could not be tested because the regression estimation algorithm would not converge § Clear determinations could not be made from a post-RCT radiograph in terms of the presence of recurrent decay, quality of the endodontic fill, or whether the restoration had an overhang or open marginStatistical methods linear mixed models (GLIMMIX ated using an analogous multipleAll analyses were done using SAS procedure) were used to implement linear regression model, also imple-(SAS Institute, Inc.). Statistical logistic regression analyses of these mented with GLIMMIX in order tosignificance refers to probabilities outcomes (extractions, clinical fail- account for correlated observations.of less than 0.05. To account for ure, and radiographic failure) (see The GLIMMIX procedure imple-the fact that individual RCTs were Tables 1 and 2). Associations with ments models (including randomclustered within individual patients time to failure (that is, associations effects) to account for correlatedand dental practices (which might between the predictor variables and observations. A pseudo-likelihoodcorrelate outcomes), generalized the outcome of interest) were evalu- estimation approach is used to fit www.agd.org General Dentistry January/February 2010 31
  5. 5. Endodontics  Outcomes of root canal treatment in DPBRN practices Table 2. Results from three multiple logistic regressions in which three different measures of failure were used as the outcomes for analysis. Odds ratio Generalized χ2 Outcome Explanatory variable (95% Cl) p value Regression (per df) Tooth was extracted Permanent restoration 3.7 (1.2; 11.5) 0.03 170 128.2 (0.79) Tooth was extracted or received endodontic retreatment Permanent restoration 4.1 (1.4; 12.2) 0.01 170 138.5 (0.85) Tooth had radiographic evidence of failure Permanent restoration 6.1 (1.2; 31.4) 0.03 113 104.5 (0.99)statistical models to data that are their patient contact time perform- varied widely as well; five practicesnot necessarily normally distributed. ing restorative dentistry procedures responded that 21–30% of revenuesThis approach allows the models that are not related to dental were derived from self-pay; two saidto account for correlations among implants (for example, amalgams, 31–40%; three said 41–50%; twoobservations—which may occur composites, crowns, bridges, posts, said 71–80%; and one said thatwhen multiple observations are foundations); one practice reported 81–90% of revenues were derivedmade on the same sampling unit that the figure was closer to 90%. from self-pay.(that is, multiple teeth within Practitioner-investigators reportedthe same patient)—and for non- that the median waiting time for Characteristics of theconstant variability among groups. a treatment procedure appoint- RCT-treated patients Using t-tests, the mean ages of ment was five days (range = 2–30 Among the patients who receivedpatients with and without clinical days). When asked to characterize RCT, 55% were female. A total offailure of teeth and those with and their practice workload during 94% of patients who received RCTwithout radiographic evidence of the previous 12 months, three were non-Hispanic Caucasian;failure were compared. practitioner-investigators reported the remainder were non-Hispanic that their practice was “too busy to African-American. Patients ranged inResults treat all people requesting appoint- age from 18–85, with a mean age ofCharacteristics of the practices ments”; four responded that they 49.2 (SD ± 12.9) at the time of RCT.All 13 of the practitioner-investiga- “provided care to all who requestedtors who participated had been in appointments, but the practice was Characteristics of thegeneral practice from 11–39 years; overburdened”; four responded RCT-treated teeth92% (n = 12) were male; 85% (n = that they “provided care to all who Of the 174 teeth that received RCT,11) were non-Hispanic Caucasian, requested appointments, and the 17% (n = 30) were anterior teeth,while the remaining 15% (n = practice was not overburdened”; and 33% (n = 57) were premolars, and2) were non-Hispanic African- two responded that they were “not 50% (n = 87) were molars. For 44%American. Among the practitioner- busy enough—the practice could (n = 77) of all teeth, the initial RCTinvestigators who participated in the have treated more patients.” was accessed by going through anstudy, 23% (n = 3) devoted 1–10% The racial mix of patients in these existing prosthetic crown. At theof patient contact time to perform- practices varied widely. One practice time RCT was initiated, 3% (n =ing RCT procedures, while 8% (n reported that more than 90% of 5) of teeth served as abutments for= 1) spent 11–20% of their time patients were black or African- RPDs; 6% (n = 11) were fixed par-on these procedures; the remaining American, while three practices tial denture abutments at the time69% (n = 9) devoted no time to reported less than 10%. The per- that the RCT was initiated. Therethese performing these procedures. centage of practice revenues derived were no overdenture abutments.All of the practitioner-investigators from patient self-payments (rather Of the 115 teeth with radio-reported spending at least 51% of than a dental insurance source) graphic data, 64% (n = 74) had32 January/February 2010 General Dentistry www.agd.org
  6. 6. two proximal contacts at the time permanent restoration was 215.4 of failure (that is, extraction orof initial RCT access, while 31% (609.1), with a range of 0–6,289 endodontic re-treatment). Among(n = 36) had one proximal contact days. When two outliers were these teeth with clinical failure, theand 4% (n = 5) had none. Of excluded from the calculation (one mean (SD) clinical survival timethese 115 teeth, 16% (n = 18) had tooth that went 2,210 days before was 7.3 (5.0) years, with a range ofradiographic evidence of dental restoration and another that went 0–18.0 years.caries (on the pre-RCT radiograph) 6,289 days), the mean (SD) number Among those teeth that hadin the tooth area that was accessed of days to permanent restoration preoperative and postoperativefor RCT. Twenty-three percent (n = was reduced to 158.5 (285.7). radiographs of satisfactory quality (n26) of these teeth with radiographic Based on the results of the survey = 115), 30% (n = 35) were judgeddata had a pre-RCT PAI score of 1; (after excluding the two outliers), as having radiographic evidence of29% (n = 33) had a score of 2; 32% permanent restorations were com- failure; these teeth had a mean (SD)(n = 37) had a score of 3; 15% (n = pleted sooner when RCT was done survival time of 7.3 (4.2) years, with17) had a score of 4; and 2% (n = 2) by an endodontist (mean (SD) of a range of 0.8–21.8 years. Amonghad a score of 5. 140.8 (239.1) days) than when RCT all teeth with any data, 27% (n = The mean (SD) percentage of was done by a general dentist (mean 47) were extracted, failed clinically,bone loss at the time of the RCT (SD) of 177.8 (329.9) days), but and/or demonstrated radiographicwas 16.7 (7.8), with a range of this difference was not statistically evidence of failure.0.5–41. At the time of the post- significant. Of the 81 teeth treatedRCT radiograph, the mean percent- by general dentists, 22 received RCT Characteristics related toage had increased to 17.8 (8.4). in one appointment, compared to 80 failure rateThe difference in bone loss between of the 93 teeth treated by endodon- Using three definitions of treat-pre- and post-RCT was not statisti- tists. This difference was statistically ment failure, the characteristics ofcally significant. significant (p < 0.05). patients and teeth were analyzed Endodontists performed RCT on Based on the post-RCT radio- to quantify their relationships to52% of teeth; the remaining 48% graphs of teeth that had not been the probability of RCT failure (seewere done by general dentists. This extracted, 7% of the teeth had Table 1). Neither patient gender orvaried by tooth type. Thirty percent recurrent caries and 2% of the res- race were significantly associatedof anterior tooth RCTs were done torations were judged to be poor by with failure. The patient’s age wasby endodontists, compared to 51% the evaluators. significantly associated with clinicalof premolars and 60% of molars (χ2 and radiographic failure but nottest; 2 df; value = 7.94; p < 0.02). Survival time and failure rate with extraction. The mean (SD) Based on the post-RCT radio- of RCT-treated teeth age of those with one or more teethgraph of teeth that had not been Among all 174 teeth, the mean showing clinical failure was 50.5extracted, the evaluators judged that (SD) length of the period from (11.8), compared to 52.4 (13.0)15% of teeth (n = 2) had a poor initial treatment to follow-up was for those without clinical failureRCT restoration (that is, with an 8.6 (5.1) years. When limited to (p < 0.05; t-test). The mean ageoverhang or an open margin). At those teeth that were not extracted (SD) of those who had one or morethe time that RCT was initiated, and showed no evidence of clinical teeth with radiographic evidence ofonly 1% of patients received a pre- or radiographic failure, the mean failure was 49.7 (10.6), comparedscription for antibiotics. (SD) length of the follow-up period to 54.5 (13.0) for patients with no was 8.8 (5.1) years, with a range of radiographic evidence of failure (p <Treatment received after RCT 1.2–28.6 years. 0.05; t-test).A total of 89% of teeth (n = 154) A total of 18% of teeth were The receipt of a permanent resto-ultimately received a permanent extracted at some point after RCT ration was a statistically significantrestoration; in 50% of these cases had been completed. Among these predictor of failure (that is, patients(n = 77), the restoration was placed extracted teeth, the mean (SD) time who did not receive a permanentwithin 37 days of the completion from RCT to extraction was 6.7 restoration were more likely to expe-of RCT. Among those teeth that (4.5) years, with a range of 0.3–16.8 rience RCT failure), as was whetherreceived a permanent restoration, years. Twenty percent of teeth were RCT was performed on a tooththe mean (SD) number of days to judged as having clinical evidence that also served as an abutment for www.agd.org General Dentistry January/February 2010 33
  7. 7. Endodontics  Outcomes of root canal treatment in DPBRN practicesan RPD at the time that RCT was those found in the literature. A the failure rate was substantial. Hadinitiated. Radiographically, two fac- success rate of 90–95% is com- radiographs been excluded from thetors were associated with failure: the monly cited as a clinical expectation study, the failure rate would havenumber of proximal contacts at the and some studies have observed been 20% and the delayed placementtime RCT was initiated and whether success rates at this level.16 A of a permanent restoration wouldthe endodontic fill was judged to be systematic review of the literature have remained the key predictor ofpoor (see Table 1). from 1966–2004 identified 306 failure. The results from this study articles that involved the outcomes justify additional investigation inResults from the multiple of RCT, 51 of which included community-based practices. Such anlogistic regression at least 100 teeth.17 The review investigation should have a prospec-Table 2 shows the results of three observed a radiographic success rate tive study design and full measure-logistic regressions that used three of approximately 82% over a five- ments should be obtained using alldifferent outcomes for analysis. All year follow-up period; during that three definitions of failure.regressions were adjusted for subject time, approximately 95% of teethage, gender, race, and type of tooth, remained functional (that is, they Definitions of treatmentalthough these variables were not were not extracted).17 successstatistically significant. The sample size in the present The literature has made many One variable (whether the tooth study exceeded the “100 teeth” attempts to define success of endo-received a permanent restoration) threshold used in the 2004 literature dontic treatment.18 Some definitionswas statistically significant and had review.17 Although the small sample have relied on tooth fracture,a substantive effect size (large odds size in this study was a limitation, periapical changes determined fromratio), regardless of which defini- a key advantage of this study was radiographs, the need for retreat-tion of failure was used. The point that it used a community-based ment, or the receipt of retreatment.estimates for its odds ratios ranged sample of practices. Traditionally, The clinical significance of thesefrom 3.7–6.1, depending on the clinical research studies have been changes and their effect on sub-outcome used for analysis. conducted only in academic settings sequent treatment have not been Other variables (not shown in (whose dentists, patients, and clini- without controversy.Table 2) also were statistically signif- cal circumstances may not be appli- Some studies have used toothicant for some definitions of failure. cable outside of those settings) or in retention as a key measure of treat-Whether the endodontically treated settings that included only patients ment success.1,2,19-21 Clearly, a majortooth was an abutment for an RPD with dental insurance. The sample justification for much of dentalpreoperatively was significantly asso- of “real-world” general dentistry care (including RCT) is toothciated with failure due to extraction practices in the present study—with retention; using tooth retention as aand with clinical failure (extraction RCTs performed by general dentists key outcome measure is consistentor endodontic retreatment), with and endodontists and the inclusion with studies that have used it as apermanent restoration, subject age, of both insured and uninsured measure of quality of care.22subject gender, subject race, and patients—suggests a composite fail- Although the need to extract antooth type also in the regression. ure rate (either extraction, clinical endodontically treated tooth mayAdditionally, whether the endodon- failure, or radiographic failure) of not be due directly to the RCT,tically treated tooth had at least one approximately 27% during a mean RCT nonetheless is the precipitatingproximal contact preoperatively was follow-up period of more than eight event leading to a chain of restor-significantly associated with radio- years after RCT. ative treatments that themselvesgraphic evidence of failure, with Additional limitations in the pres- can act as direct causes of treatmentpermanent restoration, subject age, ent study included the retrospective failure. Approximately 18% of thesubject gender, subject race, and nature of the design, the need to teeth in this study were extracted attooth type also in the regression. rely on panoramic radiographs for some point following RCT. many of the postoperative assess- A study by Lazarski et al foundDiscussion ments, and the fact that complete a 4% overall incidence of extrac-Treatment success rates sets of radiographs often were not tion following RCT after a shorterThe present study suggests a lower available. Nonetheless, the main mean follow-up period of 14.7percentage of RCT success than conclusion of this study was that months.2 This lower rate may34 January/February 2010 General Dentistry www.agd.org
  8. 8. be due to the fact that the study Among the 89% of teeth in the those found in studies based eitherpopulation consisted entirely of current study that actually received in academic environments or amongsubjects who had dental insurance. a permanent restoration at some populations with dental insurance.A large U.S. study with a follow-up point during follow-up, the mean It also suggests that any delayperiod similar to that of the present time to restoration was 215.4 days between RCT and the restorationstudy (approximately eight years) after RCT. This finding indicates the of an endodontically treated toothconsisted exclusively of patients value of conducting observational could lead to failure and that thewith continuous dental insurance community-based studies and the restoration of an endodonticallycoverage and observed a 3% tooth value of including all teeth that treated tooth is a key predictor ofloss during that time.1 A large 2007 receive RCT, rather than limiting the treatment’s ultimate success.study from Taiwan (also limited the study to those that receive These factors should be consideredto persons with dental insurance) permanent restorations. The time to when RCT is considered in com-utilized more than 1.5 million teeth restoration after RCT was much less munity settings.and observed a tooth loss rate of 7% than ideal and was a key predictorafter a five-year follow-up period.4 for retention of the endodontically Acknowledgements By comparison, in a community- treated tooth. This investigation was supportedbased study by Tilashalski et al, only The present study used popula- by National Institutes of Health36% of patients had dental insur- tions with less dental insurance grants R21-DE-16033, U01-DE-ance. The authors observed a higher coverage and located in a commu- 16746, and U01-DE-16747. Thetooth loss rate; 19% of endodonti- nity-based setting. Based on the study protocol received approvalcally treated teeth were extracted results, the possibility of higher fail- by the Institutional Review Boardafter a mean follow-up period of ure rates should enter into treatment for Human Use at the Universityonly 24.8 months. This study also planning discussions when RCT is of Alabama at Birmingham. Theconfirmed that RCT was not per- considered for these populations. authors thank Jacqueline Love, AS,formed on teeth that had substantial Other investigators have observed CDA, who served as research assis-periodontal attachment loss.19 a higher failure rate among endodon- tant for this study. One explanation for the higher tically treated teeth that were abut-failure rate in the current study is ments for RPDs, with differences in Disclaimerthe fact that 11% of endodontically failure rates varying, depending on The opinions and assertions con-treated teeth never received whether the RPD was retained by a tained herein are those of the authorsrestorative treatment by the end of clasp or some other design.28,29 and are not to be construed as neces-the follow-up period. Temporary While periodontal bone loss is a sarily representing the views of therestorations and obturated canal key factor when considering RCT respective universities or the Nationalsystems are not impervious to (and has been associated with an Institutes of Health. The informedbacteria and coronal leakage is increased risk for tooth loss), the consent of all human subjects whoan important factor in treatment present study found no association participated in this investigation wasfailure.23 Many previous studies between tooth extraction following obtained after the nature of the pro-excluded from enrollment or RCT and preoperative bone loss.30 It cedures had been explained fully.analysis all teeth that had not is reassuring that none of the teethreceived definitive restorations. in the current study had severe bone Author informationAccording to the literature, failure loss (the mean bone loss was only Dr. Gilbert is a professor and chair,to place an adequate coronal 16%). This finding is consistent Department of General Dentalrestoration results in a lower with the notion that by avoiding Sciences, School of Dentistry, Uni-success rate.2,24-26 However, in a teeth with severe periodontal bone versity of Alabama at Birmingham,2007 study, clinicians deliberately loss, the treatment plans for RCT in where Drs. Tilashalski and Litakerchose not to restore the tooth community practices are appropri- are associate professors and Dr.based on the patient’s preoperative ate in that regard. McNeal is a research statistician.endodontic status, which suggests Dr. Boykin is a staff dentist, Dentalconfounding between preoperative Conclusion Service, Department of Veteransperiapical status and whether a This practice-based study suggests Affairs, Tuskegee, AL. Dr. Kessler ispermanent restoration is placed.27 a higher failure rate for RCT than in private practice in Fairfield, AL. www.agd.org General Dentistry January/February 2010 35
  9. 9. Endodontics  Outcomes of root canal treatment in DPBRN practicesReferences ative Group. The creation and development of of a risk adjustment model. Med Care 2003; 1. Salehrabi R, Rotstein I. Endodontic treatment the Dental Practice-Based Research Network. J 41(8):937-949. outcomes in a large patient population in the Am Dent Assoc 2008;139(1):74-81. 23. Hoen MM, Pink FE. Contemporary endodontic USA: An epidemiologic study. J Endod 2004;30 13. Makhija SK, Gilbert GH, Rindal DB, Benjamin PL, retreatments: An analysis based on clinical (12):846-850. Richman JS, Pihlstrom DJ, Qvist V; DPBRN Col- treatment findings. J Endod 2002;28(12):834- 2. Lazarski MP, Walker WA 3rd, Flores CM, Schind- laborative Group. Dentists in practice-based re- 836. ler WG, Hargreaves KM. Epidemiological evalua- search networks have much in common with 24. Friedman S, Abitbol S, Lawrence HP. Treatment tion of the outcomes of nonsurgical root canal dentists at large: Evidence from the Dental outcome in endodontics: The Toronto Study. treatment in a large cohort of insured dental PBRN. Gen Dent 2009;57(3):270-275. Phase 1: Initial treatment. J Endod 2003;29(12): patient. J Endod 2001;27(12):791-796. 14. American Dental Association. CDT 2005: Cur- 787-793. 3. Imura N, Pinheiro ET, Gomes BP, Zaia AA, Ferraz rent dental terminology, ed. 5. American Dental 25. Lynch CD, Burke FM, Ni Ríordain R, Hannigan A. CC, Souza-Filho FJ. The outcome of endodontic Association, Chicago, 2004. The influence of coronal restoration type on the treatment: A retrospective study of 2000 cases 15. Orstavik D. Reliability of the periapical index survival of endodontically treated teeth. Eur J performed by a specialist. J Endod 2007;33(11): scoring system. Scand J Dent Res 1988;96(2): Prosthodont Restor Dent 2004;12(4):171-176. 1278-1282. 108-111. 26. Mindiola MJ, Mickel AK, Sami C, Jones JJ, Lalu- 4. Chen SC, Chueh LH, Hsiao CK, Tsai MY, Ho SC, 16. Torabinejad M, Goodacre CJ. Endodontic or den- mandier JA, Nelson SS. Endodontic treatment in Chiang CP. An epidemiologic study of tooth re- tal implant therapy: The factors affecting treat- an American Indian population: A 10-year retro- tention after nonsurgical endodontic treatment ment planning. J Am Dent Assoc 2006;137(7): spective study. J Endod 2006;32(9):828-832. in a large population in Taiwan. J Endod 2007; 973-977. 27. Chugal NM, Clive JM, Spangberg LS. Endodontic 33(3):226-229. 17. Torabinejad M, Kutsenko D, Machnick TK, Ismail treatment outcome: Effect of the permanent 5. Torabinejad M, Sigurdsson A. Evaluation of en- A, Newton CW. Levels of evidence for the out- restoration. Oral Surg Oral Med Oral Pathol Oral dodontic outcomes. In: Torabinejad M, Walton come of nonsurgical endodontic treatment. J Radiol Endod 2007;104(4):576-582. RE, eds. Endodontics: Principles and practice, ed. Endod 2005;31(9):637-646. 28. Wegner PK, Freitag S, Kern M. Survival rate of 4. St. Louis: Saunders;2009:379-381. 18. Torabinejad M, Anderson P, Bader J, Brown LJ, endodontically treated teeth with posts after 6. Green LA, Hickner J. A short history of primary Chen LH, Goodacre CJ, Kattadiyil MT, Kutsenko prosthetic restoration. J Endod 2006;32(10): care practice-based research networks: From D, Lozada J, Patel R, Petersen F, Puterman I, 928-931. concept to essential research laboratories. J Am White SN. Outcomes of root canal treatment 29. Sorensen JA, Martinoff JT. Endodontically treat- Board Fam Med 2006;19(1):1-10. and restoration, implant-supported single ed teeth as abutments. J Pros Dent 1985;53 7. Green LA, Dovey SM. Practice-based primary crowns, fixed partial dentures, and extraction (5):631-636. care search networks. They work and they are without replacement: A systematic review. J 30. Gilbert GH, Shelton BJ, Chavers LS, Bradford EH ready for full development and support. Br Med Prosthet Dent 2007;98(4):285-311. Jr. Predicting tooth loss during a population- J 2001;322(7286):567-568. 19. Tilashalski KR, Gilbert GH, Boykin MJ, Shelton based study: Role of attachment loss in the 8. Lenfant C. Clinical research to clinical practice: BJ. Root canal treatment in a population-based presence of other dental conditions. J Periodon- Lost in translation? New Engl J Med 2003; adult sample: Status of teeth following endo- tol 2002;73(12):1427-1436. 349(9): 868-874. dontic treatment. J Endod 2004;30:577-581. 9. Mold JW, Peterson KA. Primary care practice- 20. Aquilino SA, Caplan DJ. Relationship between based research networks: Working at the inter- crown placement and the survival of endodonti- Manufacturers face between research and quality improvement. cally treated teeth. J Prosthet Dent 2002;87(3): SAS Institute, Inc., Cary, NC Ann Fam Med 2005;3 Suppl 1:S12-S20. 256-263. 800.727.0025, www.sas.com10. Tabak LA. Dentistry on the road(map). J Am 21. Tickle M, Milsom K, Qualtrough A, Blinkhorn F, Dent Assoc 2004;135(10):1362-1366. Aggarwal VR. The failure rate of NHS funded11. Green LA, Fryer GE Jr, Yawn BP, Lainer D, Dovey molar endodontic treatment delivered in gener- Published with permission by the Academy of SM. The ecology of medical care revisited. N al dental practice. Br Dent J 2008;204(5):E8. General Dentistry. © Copyright 2010 by the Engl J Med 2001;344(26):2021-2025. 22. Jones JA, Boehmer U, Berlowitz DR, Christian- Academy of General Dentistry. All rights reserved.12. Gilbert GH, Williams OD, Rindal DB, Pihlstrom sen CL, Pitman A, Kressin NR. Tooth retention as DJ, Benjamin PL, Wallace MC; DPBRN Collabor- an indicator of quality dental care: Development36 January/February 2010 General Dentistry www.agd.org