Survival of Complete Coverage Tooth-Retained Fixed Lithium Disilicate Prostheses: A Systematic Review

Background and objectives: Porcelain-fused-to-metal (PFM) prostheses are considered the gold standard for the replacement of missing teeth, however, these have several drawbacks. Therefore, lithium disilicate (LDS) prostheses have been introduced for the construction of fixed crowns and bridges. The aim of this systematic review was to ascertain the long-term survival of LDS fixed prostheses in comparison to other materials. Materials and methods: The focused question was ‘In patients who have undergone prosthodontic treatment (participants), what are the overall survival rate of lithium disilicate (LDS) crowns and fixed bridges; and how do they relate to survival rates of non-LDS similar restoration are the survival and com-plication rates (outcomes) of LDS-based fixed prostheses with complete coverage (intervention) higher or lower when compared to non-LDS materials (controls)?’. An electronic search was conducted in PubMED/Medline, EMBASE, Google Scholar, and ClinicalTrials.gov for articles published between January 2006 and August 2022 using appropriate MeSH terms and keywords. The following types of studies were included: (1) All types of prospective clinical studies; (2) Clinical studies focusing on the survival of fixed LDS bridges and crowns; (3) Studies using natural teeth with complete coverage as abutment for fixed LDS bridges and crowns; and (4) Studies in English. The following studies were excluded: (1) Laboratory/in vitro studies and studies on LDS prostheses with no description of outcomes or survival rates; (2) Commentaries; (3) Letters to the editor; (4) Reviews; and (5) Internal data from manufacturers. The data from included studies were extracted and the risk of bias was assessed within the studies using ROBINS-I. Results: A total of 25 studies were included in this systematic review. The overall 5-year and 10-year survival rates were of 95–100% and 71.4–100%, respectively. Generally, three-unit bridges had a significantly lower survival rate over 5 and 10 years compared to single-unit crowns. Overall, the risk of bias in the included studies was moderate. Conclusions: The LDS-based complete coverage prostheses have a survival rate ranging between 48.6% and 100%. Furthermore, due to the lack of comparative studies, the long-term function and survival of LDS prostheses compared to other material prosthesis (PFM and ZrO) is debatable.


Introduction
Porcelain-fused-to-metal (PFM) prostheses have conventionally been the treatment of choice for missing teeth, primarily due to their commendable mechanical and biological properties. Although PFM prostheses have shown 10-year survival rates of more than 95% [1], they have several limitations. Firstly, their esthetics may be compromised due to the visibility of the metal framework at gingival margins or the opaque ceramic layer that is required to mask the metal substructure [2]. Therefore, more recently, ceramiconly crowns and bridges have gained popularity. Although all-ceramic prostheses do significantly overcome the esthetic limitations of PFM crowns and bridges, they possess numerous mechanical limitations and inadequacies in their physical strength. Porcelain is traditionally a fragile material and hence fractures easily in thin cross-sections [3]. To overcome these shortcomings, stronger materials such as polycrystalline zirconia and lithium disilicate (LDS) glass ceramics have been introduced as core and surface materials for dental prostheses [4]. Nevertheless, the zirconia substructure can still result in an opaque crown if used underneath porcelain [5].
Lithium disilicate (Li 2 Si 2 O 5 ; LDS) is a relatively newer material that has been used in the manufacture of fixed dental prostheses [6]. Being a glass-ceramic, it is machinable, has excellent translucency, and possesses exceptional mechanical properties. Therefore, recent research has focused on using LDS as a material for the manufacture of pressed and CAD/CAM dental prostheses and it has been suggested that LDS prostheses perform better than zirconia [7]. IPS Empress II (Ivoclar Vivadent, Schaan, Liechtenstein), an LDSbased prostheses system that has gained popularity, was introduced as a successor to IPS Empress I (leucite-based ceramic) (6). The LDS-based IPS Empress II had three times better flexural strength than leucite ceramic and was indicated for inlay, onlay, crowns, veneers and anterior three-unit bridges [7][8][9][10][11][12][13]. In the last decade, further development and research has resulted in the formulation of newer LD ceramics (IPS Emax Press & IPS Emax CAD, Ivoclar Vivadent), showing improvements in physical properties and translucency [6,14].
For a prosthesis to be successful, it should not only restore esthetics, but it should provide masticatory efficiency without failure. A prostheses may fail due to the chipping of the layering veneer, debonding from the tooth structure or flexural fractures [8]. In a recent study, LDS prostheses were indicated to possess a 10-year survival rate of more than 90% [9,10]. Nevertheless, in another study, the 15-year survival of LDS prostheses was recorded as approximately 49% [10]. As these studies have shown controversial findings in terms of survival rates for LDS ceramic restorations [9,10], it is critical to assess the survival rates of LDS ceramics by summarizing the clinical survival rates data in a systematic review as well as evaluate the factors influencing the survival of these restorations. Therefore, the aims of this systematic review were to summarize the overall survival of fixed LDS dental prostheses and to critically appraise the literature that focuses on their survival rate.

Focused Question
A focused question was constructed in accordance with the Participants, Intervention, Control and Outcomes (PICO) protocol provided in the Preferred Reporting Items for Systemic Reviews and Meta-analysis (PRISMA) guidelines [11]. The focused question was: 'In patients who have undergone prosthodontic treatment (participants), are the survival and complication rates (outcomes) of LDS-based fixed prostheses with complete coverage (intervention) higher or lower when compared to non-LDS materials (controls)?'

Literature Selection Criteria
Prior to commencing the literature search, the investigators agreed on inclusion and exclusion criteria for the literature. The following types of studies were included: (1) All types of prospective clinical studies; (2) Clinical studies focusing on the survival of fixed LDS bridges and crowns; (3) Studies using natural teeth with complete coverage as abutment for fixed LDS bridges and crowns; and (4) Studies in English. The following studies were excluded: (1) Studies on LDS prostheses with no description of the outcomes or survival rates; (2) Commentaries; (3) Letters to the editor; and (4) Reviews; and (5) Internal data from manufacturers.

sis) [MeSH] OR (fixed prosthesis) [MeSH] OR (restoration)) [MeSH] AND (survival) AND
(failure). Following the primary literature search, any irrelevant articles were excluded based on titles and abstracts. The full texts of articles, which had the potential to be included in the review, were downloaded. Furthermore, a hand-search of the following journals was performed: Journal of Prosthodontic Research, International of Prosthodontics, Journal of Prosthetic Dentistry, Dental Materials, Journal of Esthetic and Restorative Dentistry, and Journal of Prosthodontics. Additionally, the reference lists of the downloaded full-texts were scanned to find any additional articles meeting our inclusion criteria. The 'gray literature' was searched with the assistance of the library services at King Saud University and via the filters and limitations on Google and duplicate studies were eliminated. The search was carried out by two investigators (A.A.M. and A.A.D.) independently and an inter-examiner reliability score (κ) was calculated. Any disagreements were solved by discussion. The literature search strategy is summarized in Figure 1.

Literature Search
An electronic search was conducted on the following databases and registers: Pub-MED/Medline, EMBASE, Google Scholar, and ClinicalTrials.gov for articles published between January 2006 (the year in which fixed LDS prostheses were introduced) and August 2022. The following medical subject heading (MeSH) terms were used: (lithium disilicate) Additionally, the reference lists of the downloaded full-texts were scanned to find any additional articles meeting our inclusion criteria. The 'gray literature' was searched with the assistance of the library services at King Saud University and via the filters and limitations on Google and duplicate studies were eliminated. The search was carried out by two investigators (A.A.M. and A.A.D.) independently and an inter-examiner reliability score (κ) was calculated. Any disagreements were solved by discussion. The literature search strategy is summarized in Figure 1.

Data Extraction
Each investigator extracted data from the included studies corresponding to the following general categories: study design, number of patients, number of prostheses or restorations placed, number of female patients, age range and/or mean of the patients, the prostheses design and number of prostheses used in each experimental and control group (if applicable), and the maximum observation time of the study. The data were categorized and entered into a table (Table 1). Furthermore, the following characteristics specific to the prostheses were entered in another table (Table 2): abutment tooth vitality, experience of the clinicians performing the procedures, fabrication procedure, type of LDS used (monolithic or bilayered), percentage of the repairs needed during the observation period, and the overall survival rate. Any missing data were retrieved by contacting the corresponding authors of the included studies. The data extraction was independently conducted by the aforementioned investigators (A.A.M. and A.A.D.). The data were extracted on to a Microsoft Excel worksheet. Any disagreements were solved by discussion and the extracted data were independently validated by a third subject-matter expert.

Risk of Bias Assessment
The Risk of Bias in Non-Randomized Studies of Intervention (ROBINS-I) developed by Cochrane [12] was used to assess the relative levels of bias in the included studies. Briefly, the following possible sources of bias were assessed to assign each study an overall level of bias: confounding, selection, classification of interventions, deviation of intended interventions, missing data, outcomes, and selective reporting.

Factors Affecting the Survival of LD Restorations
Overall, it was observed that LDS prostheses survived better on vital teeth than on devitalized teeth [13,17]. The type of material or coverage did not have a significant impact on the survival of the prostheses [34]. Similarly, the position of the prostheses had no significant impact on the survival or complications of the prostheses [9]. It was also observed that LDS-based FDPs had a failure rate of 52.4% after 15 years [10], indicating that LDS prostheses with a longer span have a lower survival rate compared to single crowns. The majority of the fractures of the FDPs occurred at the connectors [24]. In another study, complete-coverage LDS restorations had a significantly higher survival rate compared to inlays [15].

Discussion
The results from the present study indicate that LDS complete coverage prostheses have a 5-year and 10-year survival rate of 95-100% [14,15] and 71.4-100% [22,24,30,32], respectively. Nevertheless, results by Marquardt and Strub (2006) have revealed a significantly lower 5-year survival rate for LDS-based fixed partial denture with complete coverage [14], which suggests that LDS is currently more suited to construct single crowns than prostheses with a longer span. This is most likely due to the higher flexural forces experienced by bridges relative to crowns [36]. However, this hypothesis should be considered with caution because there is a lack of comparative studies assessing the comparative survival rates of LDS crowns and bridges. Given this, one study by Hammoudi et al. revealed that the 6-year survival rate of LDS prostheses is as high as 99.7%, which was similar to Zirconia prostheses, even in patients with significant tooth-wear, which is indicative of bruxism [35].
The studies by Wolfart et al. suggested that the survival rate of LDS-based prostheses ranged between 100% after 3 years and 94% after 8 years [13,18,19]. Interestingly, in one study, they also observed that LDS failed at a significantly higher rate when resin composites were used instead of glass ionomer cements as adhesives [19]. Kern et al. [22] demonstrated a higher complication rate in resin composite-retained prostheses in comparison to glass ionomers. Similarly, in the study by Esquivel-Upshaw et al. (2008) [16], approximately 5% of LDS prostheses failed within 5 years of an observation period when resin-modified glass ionomer cements (RMGICs) and resin composites were used. Although these results seem to suggest that conventional glass ionomers are more suitable for use with LDS fixed prostheses, none of the studies included all three major classes of adhesives (conventional glass ionomers, resin-modified glass ionomers, and resin composites) to ascertain a more definitive recommendation regarding the most appropriate adhesive system to retain the prostheses. Indeed, other studies recorded 6-to 10-year survival rates of more than 90% when the resin composites were used [30,31,35] which are in contradiction with previous studies. These results warrant future comparative studies that look at the different adhesive systems to synthesize appropriate guidelines for the retention of LDS prostheses.
Since chronic bruxism has been observed to have detrimental effects on the survival rate of dental prostheses [37], these results suggest that LDS-based prostheses are a promising material for patients with parafunctional oral habits. In the included studies, it was observed that none of the variables had a significant or conclusive effect on the overall survival rate of the prostheses [9,10,. The type of fabrication (CAD/CAM or lostwax/laboratory) resulted in comparable survival rates [9,10,. Nevertheless, the reduced processing times and appointments involved in the chairside CAD/CAM fabrication of the prostheses do indeed present worthwhile advantages when compared to laboratory process dental prostheses [38]. Therefore, we hypothesized that CAD/CAM LDS-based protheses will gain popularity in the future. Moreover, LDS prostheses have been shown to have fracture/complications and survival rates similar to those of Zirconia [27,33,35]. However, to date, only three studies have compared the survival and complication rates of zirconia and LDS [27,33,35], which warrants more research to compare both the materials. Interestingly, only two studies compared the survival or complications of LDS prostheses to those of PFM crowns, considered the 'gold standard' [18,20]. In both studies, none of the LDS prostheses experienced any complications or failures, compared to 3.3% PFM prostheses experiencing complications and no failures after 3 years [18]. In the other study, however, LDS prostheses experienced a significantly higher complication rate (46%) and a lower survival rate (62.7%) compared to PFM prostheses (complication rate: 11%; and survival rate 94.7%) after 6 years [20]. The only study observing LDS-based three-unit restorations for 15 years had recorded a survival rate of 48.6% [10], which is lower than the 66.5% of PFM three-unit FDPs reported by previous studies [39]. This suggests that LDS prostheses may survive a lower rate than PFM crowns and bridges, but more studies are needed to ascertain the comparative long-term survival of both types of prostheses.
Due to the multifactorial failure and complications of dental prostheses, it is difficult to standardize the clinical studies conducted to compare or observe the performances of dental prostheses. However, it has been generally agreed that survival rates lower than 90% are considered poor [8]. Several factors impact the success (or failure) of fixed allceramic and PFM prostheses. According to Chadwick et al. [40], these factors are: the type of restoration, size and site of restorations, age, gender, socioeconomic characteristics of the patients, and the oral hygiene status. Furthermore, the age, salary, and experience of the operator also play a role in the overall outcomes of prosthodontic rehabilitation. The majority of studies included in this review did not investigate the effects of these variables on the success or complications of LDS prostheses. Only a handful of studies stated the overall duration or type of clinical experience of the practitioners involved in the clinical phases of prosthodontic rehabilitation [13,17,25,29,31], so future research could focus on the impact of operator experience on the survival of LDS prostheses. Indeed, in two studies, a 4-year survival rate of 93-95% was observed in crown preparations completed by final year students, which indicates that operator experience may not have a significant impact on the outcomes of LDS-based prostheses [29,31]. Studies indicate that the choice of adhesive has no significant overall impact on the functionality or survival of LDS prostheses [16,17,22], mirroring the outcomes observed by a previous systematic review on the survival rate of CAD/CAM-only prostheses [7]. No clear or significant differences were observed due to the tooth position or type in the studies that included these variables, which suggests that LDS-based prostheses can be used for the restoration of any type or position of tooth [9,10,. In the studies we reviewed, no clear trend was observed in the type of complications-there was equal predisposition of biological (e.g., caries, pain, gingivitis, etc.) and mechanical (fracture, porcelain fracture, etc.) complications [9,10,. However, in the majority of studies, these complications were apparent 2 years after the cementation of the crown [9,10,.
A strength of this review was that the evaluators were able to include 25 clinical studies which, to the best of our knowledge, is the first time this has been achieved. Nevertheless, due to the heterogeneity in the methodologies and study groups among the included studies, no meta-analysis was conducted in this systematic review, which can be considered a limitation of this study. Therefore, it was not possible to pool the overall outcomes and survival rates of the prostheses. Another limitation was the inability to review non-English papers because the investigators were not proficient in languages other than English or Arabic. Since none of the studies included explicit descriptions of randomization process or blinding, it is difficult to deem them internally valid. The quality assessment of the studies revealed several sources of bias which may have influenced their outcomes. Another major limitation of the included studies was that the majority of them did not look at survival rates exceeding 5 years. Only two studies recorded the complication or survival of LDS prostheses for 10 or more years. Future studies should follow up patients for longer periods of time to determine a more meaningful conclusion regarding the survival of the prostheses. Additionally, the majority of studies did not attempt to reduce the influence of other confounding variables such as parafunctional habits, smoking, or other forms of substance abuse, variables which future studies should include to evaluate their effects on LDS prostheses in comparison to other materials. None of the included studies included a cost-effectiveness evaluation of LDS prostheses so it is unknown whether the associated costs and outcomes involved in their usage is similar to or better than currently available materials. Another worthwhile avenue to look at would be the comparison between the survival rates of LDS prostheses processed by more seasoned clinicians or specialists with those made by general practitioners, fresh dental graduates, or dental students. To date, not many studies have compared the survival or complication rates of LDS with those of other materials such as porcelain, PFM, base metals, and titanium alloys. Therefore, future studies should include these comparison groups. The lack of standardization, suitable comparison groups and inadequate follow-up make the external validity of these studies debatable. Based on these limitations, the characteristics, quality, and outcomes of the included studies, it may be suggested that there is a low level of evidence that the survival and complication rates of LDS-prostheses are similar to those of other materials; further research is strongly advocated before they can be used more widely.

Conclusions
LDS-based complete coverage prostheses have survival rates ranging between 48% and 100%. Furthermore, due to the lack of comparative studies with sufficient followup, the long-term function and survival of LDS prostheses compared to other material prostheses (PFM and ZrO) is debatable.