Success rates comparison of endodontic microsurgery and single implants with comprehensive and explicit criteria: a systematic review and meta-analysis
Article information
Abstract
Objectives
While the success criteria of endodontic microsurgery (EMS) have been consistently defined and widely accepted, the success criteria of dental implants are outdated and focus only on the implant fixture and surrounding bone. This study aimed to evaluate the outcomes of EMS and single implants (SIs) with explicit criteria.
Methods
We searched for articles published from January 2010 to February 2022 and discussed them and consulted with a clinical advisory committee composed of four dental specialists and one epidemiologist during article selection and data extraction.
Results
Twenty-two EMS studies and six SI studies were included in the meta-analysis. Teeth treated using EMS had a pooled success rate of 89% (90% at <5-year follow-up and 80% at ≥5-year follow-up) and the pooled success rate of SI was 78%.
Conclusions
The success rates of the two procedures with similar follow-up periods were comparable. Subgroup analysis found no other variable that significantly influenced study heterogeneity. Considering the treatment sequence and the similar success rates, it would be advantageous to consider EMS, rather than implants, first in a situation where both procedures are applicable.
INTRODUCTION
Preservation of natural teeth is essential for oral health, and root canal treatment saves pulpally involved teeth by treating pulpal infections. When a tooth is not healed after nonsurgical root canal treatment, a surgical approach has been used to solve periradicular infection. Endodontic microsurgery (EMS) has produced reliable outcomes [1–3]; however, dental implants are frequently selected even when EMS is available. Dental implants have been popular, with economic benefits relative to the effort involved, and high success or survival rates [4–6]. Thus, as treatment options, EMS and dental implants have become not consecutive but coordinated and comparable. Some systematic review studies have compared the outcomes of the two procedures and concluded that both are valid options [7,8].
Despite the numerous complications of dental implants, most people, including dentists, expect better long-term success with dental implants compared with EMS. Outcomes of EMS have been reported as success rates, but most studies of dental implants have reported only survival rates rather than success rates. The success of EMS has been consistently defined as complete radiographic healing and the absence of clinical signs and symptoms [2,9,10]. On the other hand, success criteria for dental implants were established in 1986 when a majority of dental implants on the market were external-connection type implants (ECIs). The criteria were less than 1.0 or 1.5 mm of bone loss in the first year of service and annual bone loss of less than 0.2 mm thereafter [11–13]. Two representative internal-connection type implants (ICIs) are the Astra Tech Implant System, introduced in 1985, and the International Team for Implantology (ITI) Dental Implant System, which was introduced to the United States in 1990. ICIs have now surpassed ECIs in market share. The success criteria for ECIs are not valid for ICIs, because ICIs do not allow first-year bone loss of 1.0 or 1.5 mm, due to the so-called “platform switching effect” [14–17]. Success criteria for ICIs that do not include the parameter for bone loss have been developed [18,19]. Consistent with this, several recent studies of ICIs do not use bone loss as a success criterion but rather report just a survival rate [20–23].
Although a dental implant is composed of the implant fixture, abutment, and implant-supported prostheses and is surrounded by alveolar bone and soft tissue, the earliest implant success criteria focused only on bone loss around the implant fixture. The implant-prosthetic complex should be evaluated as a whole, and a systematic review has suggested including peri-implant soft tissues, prosthodontic parameters, and patient satisfaction as well as peri-implant bone loss in the assessment of implant success [24]. Some recent studies evaluating dental implant outcomes have incorporated various parameters but have still reported survival rates instead of success rates and additional information on complications [20,23,25]. To our knowledge, no systematic review adopting comprehensive success criteria for implant outcomes has been published.
Thus, the aim of this systematic review and meta-analysis was to evaluate the success rates of EMS and dental implants based on comprehensive criteria and to compare the success rates of the two procedures to help clinicians choose between them.
METHODS
Reporting was conducted as per the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guideline [26].
Search strategy
We searched three international (Ovid-Medline, Ovid-Embase, and Cochrane Library) and five Korean local databases for articles published from January 2010 to February 2022. The search strategy is included in Supplementary Material 1 and was developed after consultation with experienced experts to ensure its adequacy and sensitivity. In addition, the bibliographies of relevant articles were also reviewed to identify additional publications. Two independent reviewers (MK and JP) screened titles and abstracts for potential eligibility. Discordance throughout the process was discussed and further consultation with a clinical advisory committee composed of four experienced dental specialists (two endodontists, one prosthodontist, and one oral and maxillofacial surgeon) and one epidemiologist was used to attain consensus.
Inclusion and exclusion criteria
Studies fulfilling the following pre-determined inclusion criteria were considered eligible: (1) adults aged 18 years and older receiving EMS with microscopy or implant-supported single crowns (single implants, SIs); (2) randomized controlled trials (RCTs), cohort studies, case-control studies, case series with a follow-up period of ≥1 year; (3) studies that reported a success rate for EMS or SI; and (4) published original, peer-reviewed articles. EMS success was required to meet the criteria of Rud et al. [27] and Molven et al. [28]: (1) absence of clinical signs and/or symptoms, and (2) radiographic evidence of complete or incomplete healing. We restricted studies fulfilling the following three success criteria for SI: (1) peri-implant bone loss (ECI, <1 mm at the first year and <0.2 mm annually thereafter; ICI, <0.2 mm annually after crown placement or the healing period); (2) prosthetic major complication (crown fracture to be replaced, crown infraposition, or crown loss); and (3) peri-implant soft-tissue level complication (periodontal probing depth ≥5 mm and suppuration/bleeding, or fistula formation). The exclusion criteria were as follows: non-original articles or case reports, EMS without microscopy, fixed dental prostheses or any denture, fewer than 10 teeth or implants, language other than English or Korean, or duplicate publications.
Data extraction and quality assessment
We performed pilot data extraction on several studies to standardize the data extraction form and improve consistency between the reviewers. The two reviewers who conducted the study selection independently extracted data from the selected studies into a standardized form, including (1) study characteristics: authors, year of publication, study design, study period, and setting (number and location of research center); (2) study population: inclusion/exclusion criteria, number (included/dropout, tooth/implant), sex, mean age (range), sampling method, type of tooth; (3) methods: intervention, material of retrograde root filling, type of implant; (4) success rate by directly extracting the existing data or estimating the values using the existing data; and (5) follow-up length. Any disagreement between the two reviewers was resolved by rechecking the data and discussing the case further with the clinical advisory committee.
The risk of bias in the included studies was assessed by the two reviewers with the use of the Cochrane Risk of Bias tool for RCT and Risk of Bias for Nonrandomized Studies version 2.0. tool and all discrepancies were resolved by discussion with the clinical advisory committee.
Statistical analysis
Success rates by follow-up period (<5 years vs ≥5 years) were estimated for each individual study and pooled using generalized linear mixed-model random-effects meta-analysis with the “meta” and “metafor” packages in the R statistical software, version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria); this model appears to be more robust for single-proportion meta-analyses than that of the Freeman-Tukey double arcsine transformation [29]. To evaluate study heterogeneity, which refers to the variability in success rate across the primary studies, we used the Higgins I2 statistic, with I2 >50% indicating the presence of heterogeneity [30]. When there was substantial heterogeneity in success rate across studies, we conducted subgroup analyses to further explore the causes of study heterogeneity according to predefined covariates: material of retrograde root filling (mineral trioxide aggregate [MTA] vs others), continent where the study was performed (Asia vs Europe vs North America), publication year (prior to 2015 vs 2015 or later), and study design (RCT vs prospective vs retrospective). We further performed meta-regression analysis based on follow-up years, mean age, retrograde root filling (MTA proportions, %), and type of tooth (molar proportions, %). In addition, we conducted a sensitivity analysis to assess the robustness of our results. By excluding studies one by one, the overall effects of success rate with a 95% confidence interval (CI) change were estimated. In addition, publication bias was assessed through Peters’ test, which is based on a weighted linear regression of the treatment effect on the inverse of the total sample size with weights reciprocal to the variance of the average event probability [31,32]. We used two-tailed tests of significance, and p < 0.05 was considered statistically significant.
RESULTS
Study characteristics
A total of 4,363 EMS articles were identified, and 50 full texts were retrieved. Of these, 31 were excluded according to the exclusion criteria (Figure 1A). Based on scrutinizing the reference lists of relevant articles, three additional publications were added, resulting in a total number of 22 eligible articles for EMS. The electronic search rendered 11,415 SI articles. After abstract review, 113 SI articles were identified as potentially relevant for inclusion. Of these, 107 were excluded as most of them did not fulfill the three success criteria of SI. Finally, six SI articles were included in the meta-analysis (Figure 1B). No studies directly compared EMS and SI.
Flow diagram of the study selection process. (A) Endodontic microsurgery and (B) single implant. KISS, Korean Studies Information Service System; NDSL, National Digital Science Library; RISS, Research Information Sharing Service.
Table 1 presents the characteristics of the included studies for EMS. The mean age of the participants was 46 years, and seven studies were conducted in Asia [1,2,9,10,33–50]. There were three RCT studies, and the mean proportion of molar was 35.5%. Eighteen studies had a less than 5-year follow-up; the other four studies had longer follow-ups of 5, 6, 10, and 13 years. Thirteen studies used MTA only as a root-end filling material.
Characteristics of the included studies of endodontic microsurgery
All six SI studies were conducted in Europe, and there were two RCT studies (Table 2 [20,23,25,52–54]). The mean age of the study population was 37 years, and all studies had follow-ups longer than 5 years (range, 5–18 years).
Characteristics of the included studies of single implant
Risk of bias
The results of quality assessments of four EMS studies and two SI studies using the Cochrane Risk of Bias tool are summarized in Supplementary Figures 1 and 2, respectively. All EMS studies were judged as having low risk or unclear risk of bias. As two SI studies could not blind their result assessments and one was supported by a for-profit agency, they scored high risk for bias for blinding and other biases, respectively.
Success rates
Teeth treated using EMS had a pooled success rate of 90% (95% CI, 88%–92%; I2 = 58%) at <5 years of follow-up, and 80% (95% CI, 71%–86%; I2 = 0%) at ≥5 years of follow-up; the difference between the two groups was statistically significant (p < 0.01) (Figure 2A). Regarding SI, the pooled estimate of success rate was 78% (95% CI, 70%–85%; I2 = 35%) (Figure 2B). The Higgins I2 statistic indicated the presence of heterogeneity in EMS at <5 years of follow-up and in SI.
Forest plots for success. (A) Endodontic microsurgery and (B) single implant. CI, confidence interval.
To further explore the causes of study heterogeneity, we performed subgroup analysis and meta-regression among 14 EMS articles at <5 years of follow-up. We divided the participants into subgroups according to the retrograde root filling, publication year, study design, and continent. Results revealed that the pooled success rate of EMS using MTA was higher than that of EMS using other material (MTA, 90% [95% CI, 87%–93%; I2 = 44%] vs others, 88% [95% CI, 83%–92%; I2 = 56%]) for retrograde root filling, but the difference was insignificant (p = 0.25) (Figure 3A). Studies from Asia showed a lower success rate (88% [95% CI, 82%–92%; I2 = 58%]) than those from Europe (91% [95% CI, 86%–94%; I2 = 44%]) or North America (92% [95% CI, 88%–95%; I2 = 13%]), but the difference was insignificant (p = 0.12) (Figure 3B). There was no significant difference in success rates related to publication year or study design (Figure 3C, D). Moreover, despite extensive meta-regression, no other significant variable was found to influence study heterogeneity, including follow-up years, mean age, MTA proportion, and molar proportion (Table 3). Peters’ regression test for asymmetry of the included 22 EMS studies indicated no direct evidence for publication bias (p = 0.93). In our sensitivity analyses, after excluding studies one by one among 22 EMS studies, the pooled success rate was not affected (Supplementary Figure 3). Because of the small number of SI studies, we did not conduct any further analysis to assess the cause of heterogeneity.
Forest plots for subgroup analysis of success for endodontic microsurgery. (A) Retrograde root filling, (B) continent, (C) publication year, and (D) study design. CI, confidence interval; RCT, randomized controlled trial.
Meta-regression of success for endodontic microsurgery
DISCUSSION
This systematic review and meta-analysis aimed to evaluate the success rates of EMS and SI based on comprehensive and explicit criteria. The decision to preserve or extract a tooth is crucial for oral health and must be made based on the latest scientific knowledge, with explicit criteria. While the success criteria of EMS have been consistently defined and widely accepted, the success criteria of dental implants are outdated and focus only on the implant fixture and surrounding bone. Meanwhile, a systematic review of success criteria in implant dentistry has suggested a future direction for research, namely the long-term primary outcome of an implant-prosthetic complex as a whole [24]. Therefore, this study tried to evaluate and compare outcomes of EMS and SI from the comprehensive perspective of evaluating not only the bone surrounding implants but also the prosthetic parts and peri-implant soft tissue.
Success criteria for dental implants comprise four kinds of criteria—implant level, peri-implant soft tissue, prosthetic level, and patient satisfaction—and we excluded patient satisfaction. A previous systematic review on success criteria in implant dentistry included 14 single-crown implant articles according to their inclusion criteria; however, they found only one article evaluating all four kinds of success criteria [24]. We could not include the least-used parameter because of the insufficient number of articles fulfilling the criteria. Patients’ subjective perceptions and clinicians’ objective assessments are often inconsistent when used to evaluate implant-supported crowns [51]. Patient-centered parameters such as satisfaction and esthetics should be further considered in future studies.
The first success criteria for dental implants were established in 1986. The authors stated that the criteria were deeply influenced by the inherent nature of the Brånemark implant, an ECI. The main points of the criteria were bone loss in the first year of <1.5 mm and annual bone loss of <0.2 mm thereafter [11,24]. The authors acknowledged the possibility that another type of implant system would require changes in the criteria [11]. The Astra Tech Implant System and the ITI Dental Implant System are both equipped with ICIs. The proposed success criteria for ICIs were similar to those for ECI, except that a criterion of alveolar bone loss was missing. The study proposing the criteria did not include a criterion of alveolar bone loss, though it did assess the distance between the implant shoulder and the first visible bone contact [18]. Another study focusing on the radiographic evaluation of crestal bone levels of ICIs found alveolar bone loss of over 0.2 mm/year in 27 of 160 evaluated sites, including six sites with bone loss over 1.0 mm per year. However, those conditions were not regarded as failures because they did not show any correlation with the previously proposed success criteria [19]. Alveolar bone loss is the most objective parameter assessed at the implant level and should be included in the success criteria for ICIs. Therefore, we adopted the criteria of annual bone loss of 0.2 mm/year after the healing period or crown placement.
Various kinds of prosthetic complications have been reported, such as porcelain chipping, screw loosening, crown infraposition, crown fracture, and crown loss. The previous review study classified prosthetic complications as major and minor complications, depending on the availability of chairside repairs [24]. In this study, only major prosthetic complications were regarded as failures; these were crown fractures requiring replacement, severe infraposition, and crown loss. With regard to soft-tissue level, the most widely used parameters were applied to this study. These are periodontal probing depth of 5 mm, bleeding on probing/suppuration [20,25,34], and fistula formation [23].
We adopted and proposed new success criteria that considered bone loss, prosthesis, and soft tissue. Only six articles fulfilled these criteria and none of them reported success rates, while three articles reported survival rates [20,23,25], one article presented a failure rate based on implant loss [52], and the other two articles did not report any success or survival rates [53,54]. The included articles presented detailed information regarding bone loss, the prosthesis, and soft tissue, and we calculated success rates from that information. The pooled success rate of SI was 78%, ranging from 65% to 86%. In total, 293 implants were evaluated in the included six studies, with 229 implants regarded as successful. Among 64 failed implants, 18 failed due to bone loss, 23 were considered as failures because of prosthetic complications, and the other 23 were classified as soft-tissue failures. The subgroup analysis was not performed due to the insufficient number of included articles.
The success criteria for EMS, which are radiographic evidence of healing and no sign and/or symptom, have been well established and consistently applied to most EMS studies [27,28]. We therefore included enough EMS studies, despite the inclusion criteria of publication after 2010. The pooled success rate of EMS was 89%, with 22 studies having rates ranging from 76% to 97%. The subgroup analysis showed that the success rates were consistent regardless of the continent, publication year, study design, or retrograde filling material, but not follow-up years. The studies with follow-up of over 5 years had an 80% success rate, which is significantly lower than that of studies with follow-up of less than 5 years, which was 90%. This is similar to the results of a study in which the authors included EMS studies from 2002 to 2012 and revealed a 90% success rate at 2- to 4-year follow-up and an 84% success rate at 4- to 6-year follow-up [7].
In the present study, the overall success rate of EMS was 89%, which was much higher than that of SI, which was 78%. When comparing the success rates of the two procedures based only on studies with follow-ups of over 5 years, the success rate of EMS was 80% and that of SI was 78%, which is similar. These results cannot be directly compared with the results of previous systematic review studies, because most of them had taken the success rates presented in the included studies as they were reported, though the studies did not apply uniform success criteria. One study [7] addressed the fact that many different success criteria were used in that study and the most commonly used criteria were the criteria of Albrektsson et al. [11], Albrektsson and Isidor [55], and Buser et al. [18]. With this lack of clearly described success criteria, it is not possible to compare the procedures, nor to calculate the true success rate of each procedure. These ambiguous success criteria cannot be appropriate guidelines in a situation where one of the two procedures must be selected in clinical practice, where the actual prognosis must be determined. Therefore, we have defined comprehensive success criteria for EMS and SI in this study.
Dental implants have a higher survival rate and more complications than EMS [7,8]. In contrast to EMS, where only the apical portion of the tooth is treated, a dental implant replaces the entire tooth from crown to root. Therefore, the area subject to possible complications after implantation is broader than after EMS. Our analysis shows that the success rates of the two procedures are comparable with similar follow-up periods. If EMS fails, implants can be placed after tooth extraction, but EMS cannot be performed after implants fail. Considering the treatment sequence and the similar success rates, it would be advantageous for patients to consider EMS first, rather than implants, in a situation where both procedures are applicable.
CONCLUSIONS
The success rates of the two procedures with similar follow-up periods were comparable. Subgroup analysis found no other variable that significantly influenced study heterogeneity. Considering the treatment sequence and the similar success rates, it would be advantageous to consider EMS, rather than implants, first in a situation where both procedures are applicable.
Notes
CONFLICT OF INTEREST
No potential conflict of interest relevant to this article was reported.
FUNDING/SUPPORT
This study was supported by the National Evidence-based Healthcare Collaborating Agency (NECA) of Korea (NR-20-001-51). The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
AUTHOR CONTRIBUTIONS
Conceptualization: Ko MJ, Yoon JH, Kim YT, Cho SY. Formal Analysis, Investigation, Methodology, Project Administration: Ko MJ, Park JH, Lee NR, Yoon JH, Kim YT, Cho SY. Funding acquisition: Ko MJ, Cho SY. Writing - original draft: all authors; Writing - review & editing: Ko MJ, Yoon JH, Kim YT, Cho SY. All authors read and approved the final manuscript.
DATA SHARING STATEMENT
The datasets are not publicly available but are available from the corresponding author upon reasonable request.
SUPPLEMENTARY MATERIALS
Supplementary Material 1.
Search strategy
Supplementary Figure 1.
Risk of bias for endodontic microsurgery.
Supplementary Figure 2.
Risk of bias for single implant.
Supplementary Figure 3.
Sensitivity analysis for endodontic microsurgery. CI, confidence interval.