| Sign In to gain access to subscriptions and/or personal tools. |
Post Placement Affects Survival of Endodontically Treated PremolarsDipt. Scienze Odontostomatologiche, Policlinico Le Scotte, Viale Bracci, University of Siena, Siena 53100, Italy Correspondence: * corresponding author, m.ferrari{at}mclink.it
Clinical evidence is lacking regarding the influence of the amount of residual coronal dentin and of post placement on the failure risk of endodontically compromised teeth. The aim of this prospective clinical trial was to assess whether these factors significantly affect the two-year survival of restored pulpless premolars. A sample of 210 individuals provided six experimental groups of 40 premolars in need of endodontic treatment. Groups were defined based on the amount of dentin left at the coronal level. Within each group, in half of the teeth selected at random, a fiber post was inserted inside the root canal, whereas in the remaining half of the premolars, no post was placed. All teeth were covered with a crown. The Cox regression analysis revealed that post placement resulted in a significant reduction of failure risk (p < 0.001). Failure risk was increased for teeth under the "no ferrule" (p = 0.001) and "ferrule effect" conditions (p = 0.004).
Key Words: fiber post • clinical trial • luting • endodontically treated teeth • restoration
In recent years, fiber posts have been proposed as a valid alternative to metal posts in the restoration of endodontically treated teeth (Ferrari et al., 2000b; Schwartz and Robbins, 2004). The potential of fiber posts to reduce the incidence of non-retrievable root fractures in comparison with conventional cast posts has been emphasized in several in vitro studies (Akkayan and Gulmez, 2002; Fokkinga et al., 2004; Schwartz and Robbins, 2004). Also, in laboratory studies assessing fracture resistance and mode of failure of restored pulpless teeth, post-retained restorations have shown enhanced mechanical resistance and limited occurrence of fatal failures, as compared with teeth restored without posts (Fokkinga et al., 2005; Salameh et al., 2006, 2007; Sorrentino et al., 2007a,b). Retrospective (Fredriksson et al., 1998; Ferrari et al., 2000a,b, 2007) and prospective (Glazer, 2000; Mannocci et al., 2002; Malferrari et al., 2003; Monticelli et al., 2003; Naumann et al., 2005a,b; Cagidiaco et al., 2007) clinical studies have been conducted to assess the survival of endodontically treated teeth. Due to differences in study design, inclusion criteria, number of individuals studied, and observation periods, heterogeneous failure rates have been recorded, ranging from 8% for carbon fiber posts over a seven-year observation period in a retrospective study (Ferrari et al., 2007), to 12% for glass fiber posts in a two-year prospective investigation (Naumann et al., 2005a). In vivo trials have pointed out that several baseline factors—such as tooth type, position, and function in the dental arch (Naumann et al., 2005a,b), existence of proximal contacts (Caplan et al., 2002), and type of final restoration (Aquilino and Caplan, 2002)—may have an influence on the survival of restored pulpless teeth. Additionally, several studies have indicated the degree of hard tissue loss at the coronal level as a relevant factor for the clinical success of endodontically treated teeth (Naumann et al., 2005a,b), and have highlighted the importance of preserving a circumferential dentin collar of at least 2 mm in height, the so-called "ferrule effect", for improved load-bearing ability of the restored tooth (Stankiewicz and Wilson, 2002; Zhi-Yue and Yu-Xing, 2003; Akkayan, 2004; Tan et al., 2005; Pereira et al., 2006). However, no clinical study has so far specifically addressed whether and to what extent the degree of coronal tissue loss and the placement of an endocanalar post have an influence on the clinical behavior of endodontically compromised teeth. Therefore, the present study was aimed at prospectively evaluating the two-year clinical service of root-treated premolars, with various degrees of coronal tissue loss, that were restored either with or without a post and covered with a crown. The tested null hypothesis was that neither the amount of residual coronal dentin nor the placement of an endocanalar post significantly affects the two-year survival of endodontically treated and crowned premolars.
A total of 210 individuals, 93 males and 117 females, who consecutively presented at a private dental office for receiving endodontic treatment and restoration of premolars, provided six experimental groups of 40 teeth each. No more than 2 teeth for each person were considered for the study. Informed written consent was obtained from the individuals after they had received a clear explanation of the purpose of the trial, according to a protocol preliminarily reviewed and approved by the Institutional Review Board of the University of Siena, Italy. Individuals’ ages ranged from 18 to 76 yrs, with an average of 54 yrs. In total, 240 premolars, 128 maxillary and 112 mandibular, with various degrees of hard tissue loss, were included in the study. The selected teeth needed to be in occlusal function with a natural tooth, and in interproximal contact with 2 adjacent natural teeth. If the teeth had already been endodontically treated, the inclusion criteria (symptom-free root canal filling and a minimum apical seal of 4 mm, without any periapical lesion on the x-ray) had to be met. All the clinical procedures were performed by the same operator. Six experimental groups were defined as follows, based on the amount of dentin left at the coronal level after endodontic treatment and before abutment build-up:
Within each group, in half of the teeth (n = 20), a fiber post was inserted inside the root canal to provide retention for the coronal restoration (Subgroup A), whereas in the remaining half of the premolars, no endocanalar post was placed (Subgroup B). The assignment of the teeth to either subgroup was decided by a coin toss. In the premolars with 2 roots, only 1 post was placed. For all teeth, the final restoration was a single-unit metal-ceramic crown.
Clinical Procedures
In the teeth to be restored with a post (Subgroup A), at least 24 hrs after endodontic treatment, the gutta-percha was removed with Gates Glidden drills (Dentsply Maillefer) for a length of 7–8 mm, leaving at least 4 mm of intact apical seal.
We used DT Light Post posts (RTD, St. Egrève, France), choosing the post size (1, 2, or 3) that best fit the post space. The post was tried-in and consequently shortened with a diamond bur. The Prime&Bond NT Dual Cure adhesive system was used (Dentsply, Konstanz, Germany), in combination with the dual-cure resin cement Calibra (Dentsply). The post cementation procedure was performed according to the manufacturer’s instructions (Table 1 The abutment portion was built up with XFlow flowable composite and CeramX microhybrid composite (Dentsply). The crown preparation varied from a full chamfer with a bevel interproximally and lingually, to a feather finish, depending on height and thickness of the remaining dentin. Single-unit porcelain-fused-to-metal crowns were fabricated.
Evaluation Parameters
Statistical Analysis
We applied the Cox regression analysis to assess the influence on failure rate of the presence or absence of an endocanalar post, as well as of the amount of residual coronal dentin. The level of significance was set at p < 0.05.
Data were not affected by any loss to follow-up. The overall two-year survival rate of crowned endodontically treated premolars was 81.3%. For posted teeth (Subgroup A), the two-year survival rate was higher (92.5%) than for teeth restored without a post (Subgroup B, 70%, Table 2  ). In the presence of a post, no root fracture or failure of the abutment portion was recorded; all the failure events were due to post debonding, and occurred in teeth that presented with a reduced amount of residual dentin, with one wall (at most) left at the coronal level. In 2 of the post debonding cases, failure of the endodontic treatment was also observed. For the sample teeth restored without a post, 9 root fractures and 27 crown displacements were observed. The majority of crown dislodgements and all the root fractures occurred in teeth where the remaining coronal structure before abutment build-up was reduced to one residual wall at the most. All the teeth that exhibited 4 walls at the end of endodontic treatment survived the 2 yrs of clinical service, regardless of the restorativeprocedure including or omitting the placement of a post (Table 2).
For all of the Subgroup A premolars that experienced post debonding, the post was luted again, and the teeth were maintained in clinical service. As for the cases of failed endodontic treatment, all the teeth presented with asymptomatic periapical lesions. Endodontic re-treatment was performed, and the teeth were restored to clinical service. All the root fractures except one were fatal failures. Thus, in one single case, through a periodontal surgery intervention (crown lengthening), fiber post insertion, and placement of a new crown, it was possible to restore the tooth to function. Conversely, all the other fractured roots had to be extracted. The Cox regression analysis was restricted to Groups 2–6, excluding teeth presenting with 4 intact walls, that all survived regardless of the restorative procedure. The model showed that the presence of an endocanalar post was a significant factor for failure-free time. In particular, teeth restored without a post were more likely to have a shorter time to failure (hazard ratio, HR = 4.9; 95% CI for HR = 2.4 to 10.3; p < 0.001). Also, the amount of residual coronal dentin appeared to influence survival significantly. Specifically, failure risk was increased for teeth under the "no ferrule" (HR = 12.3; 95% CI for HR = 2.8 to 53.7; p = 0.001) and the "ferrule effect" conditions (HR = 8.6; 95% CI for HR = 1.9 to 38; p = 0.004). ’Post by dentin amount’ interactions were not significant (p = 0.9).
The design of this study was specifically meant to assess whether the degree of coronal tissue loss and the placement of an endocanalar post have an influence on tooth survival by standardizing, as much as possible, other variables that may affect clinical service, such as tooth type and function within the dental arch (only premolars with natural teeth as neighboring and antagonist elements), as well as type of final restoration (single-unit crowns). The first relevant finding of the investigation was that, over a two-year observation period, post placement resulted in a lower risk of failure. In particular, fiber posts appeared to contribute a protective role against what could be considered the most serious type of failure for tooth prognosis, i.e., root fracture. No posted root experienced such a breakdown. Conversely, in agreement with previous clinical reports (Monticelli et al., 2003), in the present trial, loss of retention was the most frequent unfavorable event for post-retained restorations. This finding upholds the idea that a reliable bonding to root canal dentin is not easily achieved (Pirani et al., 2005), to the point that even the contribution of friction developed by the cement-coated post along root walls should be treasured for the sake of post retention (Goracci et al., 2005; Cury et al., 2006). The establishment of a valid micromechanical intraradicular bond is indeed challenged by several factors, among which the most adverse seems to be resin shrinkage stress in the unfavorable geometric configuration of the dowel space (Tay et al., 2005). In contrast to the present investigation, a previous study on the survival of post restorations reported that the most frequent mode of failure was post fracture. The overall failure rate was also higher (Naumann et al., 2005a). However, it could be argued that the posts tested in the cited trial had been proven to have a less satisfactory fatigue resistance than the RTD post used in our investigation (Grandini et al., 2005). In a previous two-year clinical trial on root-treated premolars retaining 2 coronal walls (Monticelli et al., 2003), this system had scored a 7.7% failure rate, similar to that of the present study (7.5% for Subgroup A). With regard to the role of residual coronal dentin in restored tooth survival, several studies have suggested this role to be determinant (Stankiewicz and Wilson, 2002; Zhi-Yue and Yu-Xing, 2003; Akkayan, 2004; Naumann et al., 2005a,b; Tan et al., 2005; Pereira et al., 2006). In agreement with these findings, all post decementations in our study also occurred in teeth retaining only 1 coronal wall, the sole ferrule, or even in teeth deprived of the ferrule effect. The same could be said about fractures occurring in non-posted roots. The statistical analysis revealed that the conditions defined as "no ferrule" and "the ferrule effect" were associated with a significantly higher risk for failure. Also, Kaplan-Meier plots suggested a trend for post placement to favor tooth survival the less the coronal dentin was retained. Nevertheless, between-factor interactions were not significant according to the statistical analysis. The lack of statistical significance may be related to the fairly low overall failure rate over a relatively short follow-up period. In this regard, it should be mentioned that this study’s groups are still under observation, with the aim of collecting longer-term survival data. Also, it would be of interest to extend the investigation to other teeth, such as molars and anteriors, which were excluded from this trial for the purpose of standardization. Finally, with regard to failures of endodontic treatment, since they were all concomitant with failures of the restorative procedure, whether post decementations or crown dislodgements, it can be speculated that they were caused by re-infection of the root canal following the loss of coronal seal, as already demonstrated in previous clinical investigations (Ray and Trope, 1995; Tronstad et al., 2000). In conclusion, over a two-year observation period, post placement resulted in a significant reduction of failure risk for endodontically treated premolars. With regard to the influence of residual coronal dentin, failure risk was significantly higher for teeth that had lost all coronal walls.
Adhesive, cement, and posts used in this study were generously donated by Dentsply Italia. Authors acknowledge institutional support of staff and facilities for this study. Received for publication August 19, 2006. Revision received March 14, 2007. Accepted for publication March 27, 2007.
Journal of Dental Research, Vol. 86, No. 8,
729-734 (2007)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
