Effect of Immediate Dentin Sealing on the Bonding Performance of Indirect Restorations: A Systematic Review

The popular immediate dentin sealing (IDS) technique is used to improve the bond strength of indirect restorations. This systematic review assessed whether bond strength is affected by the type of aging conditions, bonding agents, flowable resin composites, impression materials, temporary materials, and/or resin cement used within the IDS procedure. A comprehensive database search of PubMed, Embase, Scopus, Ovid Medline, Web of Sciences, Cochrane Library, Dentistry & Oral Sciences Source, and ProQuest was carried out up to 30 January 2024 without publication year or language limitations. Only in vitro full-texts regarding the effect of IDS on bond strength were included, and the quality of their methods was assessed via a Risk of Bias (RoB) test. In total, 1023 pertinent studies were initially found, and 60 articles were selected for review after screening for the title, abstract, and full texts. IDS application improves the bond strength of indirect restorations to dentin and reduces the negative effects of temporary materials on the bond durability of final indirect restorations. Filled dentin bonding agents or combinations with flowable resin composite are preferred to protect the IDS layer from conditioning procedures.


Introduction
The traditional protocol for indirect esthetic restorations includes preparing the tooth, making an impression, and inserting a temporary restoration before fabricating and inserting the definitive restoration [1].During the temporary phase, the prepared dentin is prone to contamination and collagen degradation by temporary cement or infiltration by oral bacteria.In addition, dentin tubules exposed during tooth preparation provide a potential pathway to the pulp, which may result in postoperative sensitivity and pulpal injury [2].To mitigate these issues, studies as early as the 1990s suggested sealing freshly cut dentin surfaces with dentin bonding agents (DBA) prior to impression making [3,4].This technique, most commonly known as "immediate dentin sealing (IDS)", has also been referred to as "resin coating" [5,6], "prehybridization" [7], or "dual-bonding" [4].
Application and polymerization of DBAs through IDS can reduce the permeability of dentin by forming an interdiffusion layer, or hybrid layer, through the interpenetration of monomers into the hard tissues [8].IDS provides many advantages, including tissue conservation, improved patient comfort, reduced bacterial contamination and marginal leakage, pulpal protection, and improved bond strength [2,8,9].
One of the most important reasons for supporting IDS is its claimed positive impact on the bond strength of definitive restorations [5,8].Since bond strength is commonly considered a reliable indicator of the longevity of dental restorations [10], this positive "Immediate Dentin Sealing" OR "resin coating" OR "pre-hybridization" OR "prehybridization" OR "dual-bonding" #2 "bond strength" OR "bonding strength" #1 and #2 In addition to a free term search, a controlled vocabulary search was also conducted.However, it was discovered that subject headings similar to the keywords used were not present in the databases.Thus, the authors attempted other methods of searching, such as a reverse strategy that utilized the subject headings in articles selected from the keyword search.However, any shared subject headings between the keyword-search articles were too general and resulted in exceedingly broad searches.Finally, the authors attempted another search method, combining subject headings with keywords and/or subheadings to decrease the number of results.However, the results remained too general to be incorporated into the screening process.Therefore, with all methods exhausted, the authors decided that this present systematic review would not employ subject headings in its searches.

Eligibility Criteria
Full-text studies that pertained to the effect of IDS on the bond strength of indirect restorations to dentin and included a control group with conventional/DDS technique were added to this systematic review.Article abstracts, short communications, case reports, observational studies, reviews, and publications that pertained to other properties of IDS were excluded from the review.

Screening and Selection
The titles and abstracts of the collected studies were examined by three of the reviewers (N.T., J.H., Z.B.E.), who discussed their differences in opinions until a consensus was reached for the articles that fit the inclusion criteria.Full texts of each of these titles were then recovered and assessed for inclusion and detailed assessment of the experimental conditions.Finally, the reviewers considered the references from the chosen articles and determined the potential eligibility of articles in the references.Any disagreements between the three reviewers were settled by consulting a fourth reviewer (M.B.B.).

Data Extraction
Data was obtained from the chosen full texts and compiled on an Excel sheet by three of the reviewers (F.O., J.H., Z.B.E.).The obtained data included author names, publication year, tooth type, sample size, test method, and specific methodologies (adhesive agent, resin composite, restoration material, temporary material, conditioning method, luting cement, and aging).

Risk of Bias (RoB) Assessment
Two reviewers (J.H. and Z.B.E.) used a Risk of Bias (RoB) test to measure the methodological quality of the selected articles.Each article was evaluated based on (I) randomizing the teeth, (II) using materials in accordance with the manufacturer's instructions, (III) administering treatments with the same operator, (IV) description of the sample size calculation, (V) standardized sample preparation, (VI) blinding of the testing machine operator, and (VII) failure mode analysis.
Since most of the literature evaluated in the present study is in vitro experiments, the Cochrane RoB tool was unable to be used since it was designed for the evaluation of clinical trials.Therefore, the authors adapted a RoB methodology used in a similar review paper [14].If the authors of the study stated the parameter, the article was given a "Y" (yes) on that specific parameter; if there was no information, the article then received an "N" (no).Articles that reported a "Y" in 1-3 items were classified as having high RoB, 4-5 items as medium RoB, and 6-7 items as low RoB.

Inter-Rater Reliability (IRR)
Since the RoB assessment was performed by two reviewers independently of one another, an inter-rater reliability (IRR) test needs to be performed to determine the degree of difference between the two reviewers' designations.The IRR test was conducted using the kappa calculator on SPSS Statistics Version 19.0 (IBM, Armonk, NY, USA) following the procedure outlined in Hao et al. [15] and McHugh [16].This test calculated the percent user agreement by dividing the number of articles with the same RoB from both reviewers by the total number of articles.To run a Cohen's Kappa test, which requires the difference between the two author's designations, "Y" was converted to 1, and "N" was converted to 0, and the resulting kappa values are reported.In order to find a reliable percentage of data, the reviewers squared the kappa values from each of the parameters.Finally, using the percentages, the reviewers were able to characterize a level of agreement for each of the parameters [16].

Search and Selection
Altogether, the database, grey literature, and reference search showed 1023 pertinent articles.The flowchart of the article selection procedure, according to the PRISMA guidelines, is presented in Figure 1.After duplicate removal, the reviewers considered 699 records for their titles and abstracts.A total of 621 studies were eliminated for not adhering to the eligibility criteria, and the full texts of 78 articles were assessed.Of the 78 articles saved for more comprehensive analysis, 12 were eliminated for not including a conventional/DDS group and 6 for not being in English.One study found during the manual search in the references of the selected articles was included.Finally, 60 studies fulfilled all the selection criteria initially outlined by the reviewers and were included.The studies that had a control group of DDS technique that mimicked the clinical scenario of conventional technique or an uncoated surface were included.
outlined in Hao et al. [15] and McHugh [16].This test calculated the percent user agreement by dividing the number of articles with the same RoB from both reviewers by the total number of articles.To run a Cohen s Kappa test, which requires the difference between the two author s designations, "Y" was converted to 1, and "N" was converted to 0, and the resulting kappa values are reported.In order to find a reliable percentage of data, the reviewers squared the kappa values from each of the parameters.Finally, using the percentages, the reviewers were able to characterize a level of agreement for each of the parameters [16].

Risk of Bias (RoB) Test of the Studies in the Systematic Review
After analyzing the 60 articles for their RoB, both authors gave the majority of papers "N" in the "(III) single operator", (IV) sample size, and "(V) blinding of operator" criteria for lack of information.A total of 25 studies showed high RoB levels, 34 studies showed medium, and 1 showed low (Table 3).

Inter-Rater Reliability Results
Results from the IRR tests for each RoB parameter are shown in Table 4. Overall, the RoB parameters are above 95.00% in the percent user agreement, with the average being 99.047%.Based on Cohen's Kappa Test, the average percent of reliable data is 93.155%, which correlates to an almost perfect reliability.The average kappa value is 0.962.Parameter II, or using materials in accordance with the manufacturer's instructions, had an especially weak agreement level since the articles were variable in their degree of explanation.While the adherence to the manufacturer's instructions was explicitly stated in some, others mentioned the criteria vaguely in a table, thus resulting in inconsistencies between the two reviewers.However, all discrepancies were resolved by the two reviewers after discussion.

Discussion
Despite IDS's prevalence, procedural variability can affect indirect restoration adhesion.This variability alters bond strength between different interfaces like the restorative material and resin cement, resin cement and IDS layer, or IDS layer and dentin.The literature search of studies exploring IDS's effect on bond strength revealed heterogeneity in experimental methods and conditions.Therefore, in vitro bond strength studies were compared based on DBAs, flowable RBCs, impression materials, temporary materials, and/or resin cement.The control groups of the evaluated studies included dentin surfaces that were treated according to the DDS technique with the application of temporary and/or impression material, stored in water for a certain time, or uncoated surfaces without mimicking the conventional procedure.However, the increased use of CAD/CAM systems in recent years reduced the need for temporary materials and impressions.Therefore, it became possible to bond the indirect restorations immediately.Consequently, the comparison of bonding to an uncoated surface and IDS is as important as the comparison of conventional techniques to IDS.Overall, the literature supported IDS's ability to improve bond strength.A few showed negative or no impact [36,58,59,67], while several showed benefits contingents upon the resin cement [34,35,43,45,57], DBA [39,46,56], or flowable RBC [23,32,69].

Effect of Dentin Bonding Agents and Flowable Resin-based Composites
Since Pashley et al. [3] introduced dentin sealing in 1992, many DBAs have been used to seal freshly cut dentin.In 2005, Magne et al. [8] pioneered IDS by using filled DBAs (Optibond FL (OFL); Kerr Corporation, Orange, CA, USA) or combining unfilled DBAs and flowable RBC.Unfilled DBAs form thinner layers, and surface cleaning procedures may destroy the hybrid layer and re-expose dentin [8].Applying additional flowable RBC reduces dentin exposure risk, mitigates stress on the interface, and eliminates the oxygen inhibition layer by sealing DBA.Similarly, the original resin coating technique involves two-step self-etching with flowable RBC [5].
When two gold standard DBAs-3-step ER (OFL) and 2-step SE (Clearfil SE Bond; Kuraray Medical Inc., Tokyo, Japan)-were compared, OFL showed higher or similar bond strength [18,39,46,63].In addition, 2-step SE (Clearfil SE Bond) showed higher bond strength than 2-step ER, regardless of using flowable RBC [20,56].However, few studies that investigated different DBAs, including 3-step ER, 2-step ER, 2-step SE, and 1-step SE without flowable RBC, exhibited no benefit from IDS [36,58,59].The composition of the adhesives and the filler content may play a role in the different performance of the adhesives.Optimally filled adhesives showed increased mechanical properties of the adhesive layer and increased bond strength [77,78].An adhesive layer with increased mechanical properties may help the stress distribution and act as a shock absorber.However, high filler size and content also lead to high viscosity and reduced penetration of the adhesive into the dentin [79].Furthermore, the high viscosity of the filled adhesives would lead to pooling at the margins.
Half of the investigated studies combined DBA with flowable RBC, which enhanced bond strength more than DBA alone [20,23,32,46,69].Filled OFL improved bond strength compared to DDS, even without flowable RBC.Unfilled/lightly filled DBAs should be applied with flowable RBC in the "reinforced IDS" approach to improve µTBS to dentin [46].
Adding flowable RBC or extra adhesive layers similarly affected bond strength [42,63].However, the double application of all-in-one adhesives improved resin coating bond strength [31,75].One study investigated the effect of IDS form and thickness applied by universal adhesive and a flowable RBC on intra-cavity µTBS.Thicker IDS layers may act as a stress-breaker under cyclic load stress, and moderate and thick IDS layers' bond strength was higher than thin or no IDS [41].Therefore, combining DBA with flowable RBC is recommended.Akehashi et al. [43] compared dual-cure resin cement as the IDS material with flowable RBC and reported that combining two-step SE adhesive with dual-cure resin cement as IDS showed the closest results to the bond strength of direct restorations.

Effect of Impression Materials
Following IDS application, impression materials may interact with the outer resin layer [4], which is unpolymerized due to oxygen inhibition [8,80,81], yielding an unpolymerized layer of impression or impression material adhesion and tearing on the IDS surface [9,82].Eliminating oxygen-inhibition layers (OIL) is necessary to prevent interference with impression material setting and temporary restorative material adherence [9].Cleaning with an alcohol-soaked cotton pellet, pumice, or application of glycerin gel with additional polymerization are accepted methods to eliminate OIL [9,83,84].Although it is possible to obtain successful impressions with vinyl polysiloxane following air blocking or pumicing of the sealed surface, there is the risk of faulty impressions due to adhesion and tearing with polyether impression material [9,85].In several studies, cavity impressions were taken due to preparation type or to simulate clinical practice.However, the included studies did not evaluate the effect of impression materials on bond strength with IDS.

Effect of Temporary Materials and Conditioning Methods
Since contaminating dentin with temporary materials reduces adhesion [25,37,54], IDS should be applied before temporary restorations.However, temporary cement remnants may also contaminate the IDS surface and decrease bond strength to sealed dentin.Most studies showed IDS's benefits, regardless of temporary materials or cleaning methods.Sealed dentin surfaces can bind resin-based temporary materials due to the OIL.Therefore, retrieving and removing temporary materials may be difficult [6,18].Isolating sealed dentin with a glycerin gel [36] or petroleum jelly [18] can inhibit the interaction with resin-based temporary materials.Removal of non-eugenol temporary cement with the excavator and cleaning with alcohol may not be enough to remove cement remnants from the IDS surface and lead to a reduction in bond strength [50].Therefore, the conditioning method is important to remove temporary cement's harmful effects.
Different conditioning methods, such as airborne-particle abrasion with aluminum oxide, etching with phosphoric acid, polishing with pumice, and tribochemical silica coating, were used beneficially with IDS.With the tribochemical silica coating method, silica particles are deposited on the surface, which leads to an increase in the surface [86].However, three studies compared the effects of different conditioning methods on the IDS surface [42,59,63].Cleaning with pumice only, or with additional tribochemical silica coating of zinc-carboxylate cement [42], or temporary zinc-oxide luting cement [63] did not affect bond strength.However, the conditioning method depends on the type of IDS material type used.Tribochemical silica coating or sandblasting may remove a thin IDS layer and result in decreased bond strength [42].After conditioning, the dentin exposure risk may be reduced with filled adhesives or flowable RBC [42,87].Therefore, a thick IDS layer created with an extra adhesive layer or flowable composite when using silica-coating conditioning is recommended by the authors in clinical practice [63].

Effect of Resin Cement
Self-adhesive (SA) resin cement has gained popularity due to the reduced postoperative sensitivity and application simplicity [88] but has lower bond strength than conventional resin cement [89].Among studies indicating IDS's dependence on resin cement types, two [43,45] compared different conventional SE resin cement and three [34,35,57] compared conventional resin cement with SA resin cement.Others [37,55,62] showed IDS improved bond strength, regardless of the cement.Nevertheless, curing through restorative materials characterized by different translucency [90] may influence the overall cement conversion, thus possibly affecting bond strength to the dentin surface.In addition to resin cement, IDS improved the bond strength of dual-curing compomer cement [62] and resin composite as a luting material [55,76].
The remaining OIL on the resin coating following light curing may help the resin cement bond to the coating materials [91].However, this unpolymerized layer should be removed to prevent interference with impression materials.Therefore, resin-coated surface pretreatment is important to improve the adhesion with resin cement.

Bonding Efficiency to Different Restorative Materials and Preparation Types
A higher C-factor in inlay cavities causes polymerization contraction stresses on bond strength and reduces internal adaptation [45].However, IDS was found beneficial for Class II inlay [19,24,45], Class I inlay [23,25,26], indirect Class V restoration [40], onlay [38,41] overlay [61], and crown restoration bond strength [22,30,44,51], in addition to flat dentin surfaces.IDS application on crown preparations is more technique-sensitive due to the preparation deformation risk and adhesive pooling on the preparation shoulder.
Regarding the restorative material, IDS improved bond strength to ceramic [41,44,51,61], zirconia [62], metal [60], and resin composite materials.Conventional indirect restorations involve complicated procedure steps, while CAD/CAM restorations have several advantages.Ishii et al. [38] compared the intra-cavity bond strength of different CAD/CAM blocks with and without IDS.IDS improved the bond strength of Lava Ultimate and Vitablocks Mark II.However, resin composite blocks (Lava Ultimate; 3M ESPE; St. Paul, MN, USA, and Vita Enamic; VITA Zahnfabrik, Bad Säckingen, Germany) showed significantly higher bond strength than glass-ceramic blocks (Vitablocks Mark II; VITA Zahnfabrik, Bad Säckingen, Germany), regardless of IDS.The resin cement and surface treatments of the restorative materials play important roles in the interface.

Effect of Aging Conditions
Overall, IDS improved bond strength after aging; however, some studies that evaluated IDS stability over time [28,30,39,42,72] showed contradictory results regarding IDS's effect on interface durability.Although IDS may initially improve bond strength depending on the DBA, it did not prevent decreasing bond strength after water storage for 3 months [39] and 6 months [49].Another study reported the adhesive interface with IDS was stable after water storage for 6 months [42].It should be noted that the aging method used in some of the studies [19,26,54,57] included 1000-1500 thermocycles, which is not sufficient for aging standards [92,93].Therefore, there is still a need for long-term durability studies.
Magne et al. [9] recommended sealing dentin surfaces with a DBA immediately after tooth preparation for indirect restorations.They confirmed that the bond strengths were not changed by up to 12 weeks of elapsed time before the cementation of permanent restorations.

Study Limitations
Quantitative evaluation through meta-analysis could not be performed in this present review due to the heterogeneity of the evaluated studies' methods.Additionally, owing to the lack of standardized criteria for assessing the RoB and in vitro study quality, a previous study's methodology was adopted.Therefore, more recent in vitro and in vivo studies are needed to better analyze laboratory and clinical correlation data.

Conclusions
The evaluation of the included studies shows that IDS application improves the bond strength of indirect restorations to dentin and reduces the negative effects of temporary materials on the performance and the long-term durability of final indirect restorations.Filled DBAs or combinations with flowable RBCs are preferred to protect the IDS layer from conditioning procedures.In order to evaluate the potential relationship between bond strength data and clinical outcomes of the studies, more in vitro and clinical studies are needed.

Figure 1 .
Figure 1.The Prisma flow diagram.*: 12 did not include a control (DDS) group, 6 were unable to be accessed in English.

Figure 1 .
Figure 1.The Prisma flow diagram.*: 12 did not include a control (DDS) group, 6 were unable to be accessed in English.

Table 1 .
Search strategy used in PubMed.

Table 2 .
Summary of the studies included in the systematic review.

Table 3 .
Risk of bias.

Table 4 .
IRR values of studies in the systematic review.