Wet vs. Dry Dentin Bonding: A Systematic Review and Meta-Analysis of Adhesive Performance and Hybrid Layer Integrity

Abstract

Objective: This systematic review and meta-analysis aimed to evaluate the effects of moisture control strategies (including wet-bonding techniques, universal adhesives, and etching type) on dentin bonding performance in restorative dentistry. Methods: A comprehensive literature search was conducted across PubMed, Scopus, and Google Scholar, following PRISMA guidelines. Only in vitro and ex vivo studies comparing wet- and dry-bonding protocols, using human dentin substrates, and reporting microtensile bond strength (μTBS) were included. The data were synthesized using a random-effects meta-analysis and the methodological quality was assessed using the MINORS tool. Certainty of evidence was evaluated using the GRADE framework. Results: Nine studies met the inclusion criteria, eight of which were included in this meta-analysis. The moisture control strategies significantly influenced the bonding outcomes, with ethanol and acetone wet bonding yielding higher μTBS and enhanced hybrid layer morphology. The universal adhesives performed effectively under both moist and dry conditions, although their performance varied by the adhesive composition and solvent system. The meta-analysis revealed a statistically significant advantage for hydrated dentin (SMD = +1.20; 95% CI: 0.52 to 1.86; p < 0.001), with the moist and ethanol-treated substrates outperforming the dry and over-wet surfaces. The long-term durability was better preserved with ethanol and acetone pretreatments and the adjunctive use of chlorhexidine. Conclusions: Moisture conditions influence dentin bond strength, but modern universal adhesives show consistent bonding performance across different moisture conditions. Solvent-wet-bonding protocols, particularly with ethanol or acetone, enhance the immediate and long-term performance. While the current evidence is limited by the in vitro designs and heterogeneity, the findings demonstrate protocol flexibility and highlight strategies to optimize adhesion in clinical practice. Future clinical trials are necessary to validate these approaches under real-world conditions.

1. Introduction

Dental adhesive systems play an important role in modern restorative dentistry, enabling the bonding of composite materials to tooth structures. However, achieving durable adhesion to dentin remains challenging due to its complex composition and inherent moisture [1,2]. The failure of bonded restorations is frequently attributed to degradation of the hybrid layer, where hydrolytic and enzymatic activity compromise the stability of collagen fibrils and resin infiltration over time. Recent reviews have confirmed that hybrid layer breakdown continues to be a leading cause of restoration failure [3,4]. Lately, there has been the development of new adhesive strategies and materials aimed at improving the bond strength and longevity [5,6]. One key factor affecting dentin bonding is the surface moisture. The “wet bonding” technique was introduced to prevent the collapse of demineralized collagen fibrils and to allow for better resin infiltration [7,8]. However, controlling the optimal level of dentin moisture can be technique-sensitive [9]. Excessive moisture can dilute adhesive monomers, while over-drying can cause collagen collapse. To address this issue, universal adhesives have been developed that claim to be less sensitive to moisture variations [8,10]. These multi-mode adhesives can be used in etch-and-rinse or self-etch techniques. Most of them contain 10-MDP (10-methacryloyloxydecyl dihydrogen phosphate) monomer, which forms a stable chemical bond with hydroxyapatite [11]. More recently, solvent-wet approaches have been proposed. Acetone wet bonding has been shown to improve the immediate μTBS, reduce nanoleakage, and suppress endogenous protease activity even after thermal and enzymatic aging [12]. However, the effects of dentin moisture on universal adhesive performance remain subject to debate. Several studies have evaluated universal adhesives under different moisture conditions [10,13,14,15,16]. Some have found no significant differences in the immediate bond strength between wet and dry dentin for most of the universal adhesives tested [17,18]. Some adhesives performed better on wet dentin, while others preferred a dry substrate [19,20]. An alternative approach is the ethanol-wet-bonding technique, which uses ethanol to replace the water in a demineralized dentin matrix [21,22,23,24,25]. This allows for better infiltration of hydrophobic monomers. Studies have shown improved immediate and long-term bond strength with ethanol wet bonding compared to water wet bonding [16,26,27]. Acetone wet bonding has been proposed as another strategy to improve bonding [12,14,25]. The adhesive strategy also impacts bonding durability [28]. Etch-and-rinse adhesives remove the smear layer completely, while self-etch adhesives modify and incorporate it. This affects hybrid layer formation and stability. Some studies have suggested that etch-and-rinse can provide better long-term bonding, while others have found self-etch to be more durable [29,30], while others have reported greater durability with self-etch adhesives [31,32]. The incorporation of MMP (matrix metalloproteinases) inhibitors has been investigated as a strategy to preserve bond integrity over time. Several studies have shown reduced degradation when CHX was applied before adhesive procedures [29,33], while other reports have found mixed or negligible effects. This suggests that the benefits of CHX may depend on the adhesive type, substrate conditions, and storage protocols [6,34].

Nevertheless, consensus is hindered by the absence of standardized definitions for dentin moisture states, with terms such as moist, wet, and oversaturated often used inconsistently across studies. These inconsistencies, combined with the variability in adhesive formulations and aging protocols, have contributed to significant heterogeneity in the reported outcomes [35,36]. Clinically, wet bonding involves leaving the dentin visibly moist after etching, typically achieved via blot-drying with absorbent paper or gentle air, such that the surface remains glossy but not pooling. This method preserves the collagen architecture and promotes effective resin infiltration [37]. Over-drying, which results in a matte or chalky appearance, compromises infiltration by collapsing the collagen scaffold and leads to reduced bond strengths both immediately and over time [38,39]. Oversaturation, visible as water droplets on the surface, may dilute the adhesive monomers and similarly impair bonding. Moreover, while universal adhesive systems have shown variable sensitivity to surface wetness, the balance between moisture and dryness remains a critical factor in achieving durable dentin bonds [5,14,40]. However, critical gaps remain. The evidence on solvent wet bonding, particularly ethanol- and acetone-wet protocols, is conflicting: some studies have reported enhanced immediate bond strength and reduced nanoleakage, while others have suggested limited durability over long-term storage. These inconsistencies, together with the absence of a systematic synthesis distinguishing water-moist from solvent-wet strategies, highlight the need for clarification. The present review addresses this gap by evaluating both universal adhesives and solvent wet bonding within a single framework. By pooling the microtensile bond strength and hybrid layer outcomes, and appraising the evidence through the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) and GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) frameworks, this study provides a structured analysis that aims to synthesize the current evidence on dentin bonding strategies, focusing on moisture control techniques, universal adhesives, and alternative wet-bonding approaches. Through an examination of these interconnected factors, optimal protocols can be identified to ensure predictable and durable adhesion to dentin in clinical practice.

2. Materials and Methods

This systematic review and meta-analysis was conducted according to the PRISMA guidelines [41] and structured using the PICO framework. The completed PRISMA checklist is provided in the Supplementary Materials. The research question was as follows: “In dental restorative therapy using resin composites, does the application of wet bonding techniques, compared to traditional dry bonding protocols, result in improved bond strength and quality of the adhesive interface?” This question was formulated to address the clinical relevance of moisture control strategies in adhesive dentistry, particularly in relation to modern universal adhesives and resin composites. The PICO components of the review are detailed in

Table 1. PICO process.

PICO Element	Definition
Population (P)	Extracted human dentin from permanent teeth (primarily third molars) prepared for composite bonding.
Intervention (I)	Wet-bonding techniques, including water-wet-, ethanol-wet-, and acetone-wet-bonding protocols.
Comparison (C)	Conventional dry-bonding techniques (air-dried dentin without intentional rewetting).
Outcome (O)	Primary: Microtensile bond strength (μTBS, in MPa). Secondary: Adhesive interface quality (e.g., nanoleakage, hybrid layer integrity, and long-term durability).

. A comprehensive literature search was conducted across PubMed, Scopus, and Google Scholar to identify studies evaluating the effects of wet-bonding techniques on the bond strength and adhesive interface quality of resin composites in comparison to traditional dry-bonding protocols within the context of dental restorative therapy. The search strategy incorporated a combination of MeSH terms and free-text terms related to “wet bonding,” “resin composites,” “bond strength,” “adhesive interface”, and “dry bonding”. Boolean operators (“AND,” “OR”) were applied to combine the terms effectively. The searches were limited to studies published in English from 2014 to 2025 and filters for article type (e.g., in vitro studies) were applied when available. The reference lists of included articles were also screened for additional relevant studies. The eligible studies were required to meet the following inclusion criteria: (1) in vitro or ex vivo experimental design; (2) use of human dentin substrates; (3) comparison of at least one wet-bonding technique with a dry-bonding protocol; and (4) reporting of quantitative microtensile bond strength (μTBS) data and/or qualitative assessment of the adhesive interface. Studies were excluded if they did not include a dry-bonding comparison, failed to report bonding-related outcomes, or involved non-human models. Each study was independently appraised by four reviewers. For consistency, the moisture levels were categorized as follows: ‘dry’ = prolonged air-drying leading to collapsed collagen; ‘moist’ = blot-dried surface with visible gloss; ‘wet’ = visibly wet dentin without pooling; ‘oversaturated’ = intentional addition of excess water beyond natural hydration. In the meta-analysis, moist, wet, and oversaturated were grouped as ‘hydrated dentin’ unless otherwise specified. For the purposes of this review, moist, wet, and oversaturated dentin conditions were grouped under the broader category of “hydrated dentin” to facilitate an overall synthesis. However, solvent-wet strategies, specifically ethanol and acetone wet bonding, were considered separately, as they differ fundamentally from water-based hydration. These approaches rely on chemical dehydration and enhanced resin infiltration rather than the maintenance of natural dentin’s moisture. Because of these distinct mechanisms and their potential clinical implications, solvent-wet protocols are presented and discussed independently, ensuring that their unique role in adhesive performance is not obscured by a direct comparison with water-moist conditions. Disagreements were resolved through discussion and consensus. The searches were conducted electronically and manually, and verified by three individuals, and EndNote X9 (Clarivate, Philadelphia, PA, USA) software was used for the organization of references and the elimination of duplicates. During the preparation of this work, the authors used ChatGPT 4.5 (OpenAI, San Francisco, CA, USA) and Elicit Pro (Elicit, Oakland, CA, USA), accessed on 14 July 2025, in order to generate figures, improve the readability of the text, and refine the data extracted. The Methodological Index for Non-Randomized Studies (MINORS) was used to assess the methodological quality of all the included studies [42]. It should be noted that the MINORS tool was originally designed for non-randomized clinical studies, and its application to in vitro and ex vivo investigations has inherent limitations; nevertheless, it provides a structured and transparent framework for assessing methodological quality. The eight-item version of the tool, designed for non-comparative experimental studies, was applied. Each item was scored from 0 (not reported) to 2 (reported and adequate), with a maximum possible score of 16 per study. Given the in vitro nature of all the included studies, the ROB 2 tool, which was designed specifically for randomized controlled trials involving human participants, was not applicable. Similarly, the ROBINS-I tool and the Newcastle–Ottawa Scale (NOS), which are intended for non-randomized clinical studies, such as cohort or case–control designs, were not appropriate for application [43]. Instead, risk of bias was considered qualitatively, with reference to the methodological quality as assessed by the MINORS tool. Certainty of evidence for each reported outcome was assessed using the GRADE framework. All the included studies were experimental and conducted in vitro and, therefore, the certainty of evidence started at “low” by default. The data were extracted as the mean μTBS values (MPa) with the standard deviations and sample sizes for each group. The moist, wet, and oversaturated dentin substrates were grouped into a single “hydrated dentin” category to facilitate a pooled analysis. A meta-analytical approach was implemented using R software (version 4.3.1) and RStudio (Posit, version 2023.06.1) as the integrated development environment. The analysis employed the meta and metafor packages to perform random-effects modeling based on the DerSimonian–Laird estimator, which accounts for both within- and between-study variability. The standardized mean differences (SMDs) and their corresponding 95% confidence intervals (CIs) were calculated for the continuous outcomes (microtensile bond strength) using pooled standard deviations to ensure comparability across the studies that reported in consistent units (MPa). The heterogeneity was quantified using the I² statistic and the Cochran Q test; moderate-to-substantial heterogeneity prompted prespecified subgroup analyses based on the dentin’s moisture condition, etching strategy, and adhesive type. A sensitivity analysis was conducted to assess the influence of individual studies on the meta-analytic outcome. Additionally, forest plots were constructed to visualize the effect sizes and heterogeneity across comparisons. Although a meta-regression was considered, the limited number of included studies per subgroup precluded reliable estimation of moderator effects. All the statistical procedures adhered to the recommendations outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Version 6.4) [44].

3. Results

3.1. Selection of the Included Studies

A comprehensive literature search was conducted using PubMed, Scopus, and Google Scholar, yielding a total of 1245 records. An additional 11 records were identified through manual searches of reference lists. After the removal of duplicates, 874 records remained for screening. Based on the titles and abstracts, 845 studies were excluded as irrelevant. The full texts of 29 articles were assessed for eligibility. Of these, 20 were excluded due to unsuitable study design, inadequate outcome reporting, or methodological incompatibility. Nine studies met the inclusion criteria for the qualitative synthesis. Of these, eight studies were deemed methodologically compatible and included in the meta-analysis, while one was excluded from the quantitative synthesis due to its incompatible data structure [19] (Figure 1).

3.2. Characteristics of Included Studies

The analysis included nine studies examining various dental bonding techniques. Eight studies employed in vitro experimental designs, while one used a quasi-experimental ex vivo approach (

Table 2. Characteristics of the included studies.

Study	Study Design; Main Measurement	Sample Size	Bonding Technique	Measurement Methods
Leite et al., 2020 [45]	In vitro experimental; microtensile bond strength (TBS)	human third molars	Wet vs. dry bonding (Scotchbond Universal, Adper Single Bond 2)	Microtensile bond strength at 24 h, 1 year; t-test; Weibull analysis.
Saeed et al., 2021 [47]	In vitro experimental; microtensile bond strength	human third molars	Wet vs. dry; multiple adhesives (2-hydroxyethyl methacrylate (HEMA) containing/HEMA- free)	Microtensile bond strength at 24 h, after thermocycling; three-way analysis of variance.
Elbanna et al., 2024 [49]	In vitro experimental; microtensile bond strength	mandibular first molars (microtensile bond strength); 16 maxillary premolars (morphology)	Wet/dry; dimethyl sulfoxide (DMSO) pretreatment; Gluma Bond Universal	Microtensile bond strength at 24 h, after thermomechanical aging; three-way analysis of variance; environmental scanning electron microscopy (ESEM).
Nonato et al., 2022 [46]	In vitro experimental; microtensile bond strength	378 human molars	Dry, wet, oversaturated; three adhesives; etch-and-rinse (ER)/self-etch (SE)	Microtensile bond strength, SNU at 24 h, 1 year; four-way analysis of variance; scanning electron microscopy (SEM), Fourier-transform infrared spectroscopy (FTIR).
Choi et al., 2017 [39]	In vitro experimental; microtensile bond strength	72 human third molars	Etch-and-rinse/self-etch; 0, 5, 10 s air-drying; three adhesives	Microtensile bond strength at 24 h, two-way analysis of variance, confocal laser scanning microscopy (CLSM).
Hamouda et al., 2022 [50]	In vitro experimental; microtensile bond strength	60 human third molars	Etch-and-rinse/self-etch; dry, wet, moist; two adhesives	Microtensile bond strength at 24 h, three-way analysis of variance, scanning electron microscopy.
Caceres et al., 2020 [22]	Experimental ex vivo; microtensile bond strength	48 premolars	Water vs. ethanol wet bonding; Single Bond 2	Microtensile bond strength (immediate), t-test.
Zhao et al., 2023 [12]	In vitro experimental; microtensile bond strength	60 third molars	Water/ethanol/acetone wet bonding; Singlebond Universal	Microtensile bond strength at 24 h, after thermocycling and collagenase aging; two-way analysis of variance.
Ekambaram et al., 2014 [48]	In vitro experimental; microtensile bond strength	48 molars (caries affected)	Ethanol/water wet bonding with or without chlorhexidine; hydrophobic adhesive	Microtensile bond strength at 24 h, 12 months; three-way analysis of variance.

). The sample sizes varied considerably across the studies, ranging from 20 human third molars [45] to 378 human molars [46]. All the studies utilized microtensile bond strength (μTBS) as their primary outcome measure. The bonding strategies encompassed etch-and-rinse protocols; self-etch approaches; and various solvent-based techniques using water, ethanol, and acetone (Zhao et al., 2023 [12]). For the outcome measurements, all the studies primarily assessed the microtensile bond strength (μTBS) measured in MPa, with measurements taken at 24 h and at extended periods up to one year [12,45]. The statistical analysis methods varied across the studies, incorporating t-tests for independent samples; two-way and three-way ANOVAs; four-way analysis of variance; Weibull analysis; and post hoc tests, such as Tukey’s and Bonferroni [45,46,47]. The durability testing protocols included thermocycling, thermomechanical aging, collagenase aging, and long-term water storage for up to 12 months [12,48]. These aging methods were employed to evaluate the long-term stability and performance of the different bonding techniques and adhesive systems under investigation.

3.3. Methodological Quality Assessment

Nine studies met the inclusion criteria, comprising a range of universal adhesive systems tested under various conditions, such as etch-and-rinse versus self-etch modes, different dentin substrates, and aging durations. The methodological quality scores based on the MINORS tool are summarized in

Table 3. Quality assessment of the included studies using the MINORS criteria. Scoring: 0 = not reported; 1 = reported but inadequate; 2 = reported and adequate. Maximum possible score = 16.

Study	Year	Clearly Stated Aim	Inclusion of Consecutive Specimens	Prospective Data Collection	Endpoints Appropriate	Unbiased Assessment	Adequate Follow-Up	<5% Loss to Follow-Up	Sample Size Calculation	Total (/16)
Elbanna et al. [49]	2024	2	2	2	2	1	2	2	2	15
Caceres et al. [22]	2020	2	2	2	2	1	2	2	2	15
Choi et al. [39]	2017	2	1	1	2	1	2	2	0	11
Ekambaram et al. [48]	2014	2	1	1	2	1	2	2	0	11
Leite et al. [45]	2020	2	1	2	2	2	2	2	2	15
Hamouda et al. [50]	2022	2	2	2	2	1	2	2	1	14
Saeed et al. [47]	2021	2	2	2	2	2	2	2	2	16
Zhao et al. [12]	2023	2	2	2	2	1	2	2	1	14
Nonato et al. [46]	2022	2	2	2	2	2	2	2	2	16

. Most of the studies demonstrated a high reporting quality, with six studies scoring 14 or higher. The most frequent limitations were the absence of a prospective sample size calculation and limited reporting on blinding of outcome assessment. The lowest-scoring studies (Choi et al. and Ekambaram et al.) scored 11, primarily due to a lack of prospective planning and sample size justification. The quantitative synthesis indicated moderate heterogeneity in the outcomes, attributable to variations in the adhesive types, operator controls, storage protocols, and methodological rigor.

3.4. Certainty of Evidence (GRADE)

Certainty of evidence for all outcomes was rated as very low, reflecting the limitations associated with the indirectness, inconsistency, and imprecision inherent in in vitro research. Table 2 summarizes the GRADE assessments across the four key outcomes: microtensile bond strength, nanoleakage, hybrid layer degradation, and long-term durability. Formal assessment of publication bias was not possible due to the limited number of studies per outcome and the absence of standardized effect estimates. No study protocols were registered in advance and no unpublished data were identified despite the grey literature searches. The publication bias was therefore judged as “unclear” for all outcomes (

Table 4. GRADE assessment of the certainty of evidence for the key outcomes of this review.

Outcome	No. of Studies	Study Design	Risk of Bias	Inconsistency	Indirectness	Imprecision	Publication Bias	Certainty of Evidence
Microtensile Bond Strength (μTBS)	9	In vitro experimental	Not serious	Serious	Very serious	Serious	Unclear	Very low
Nanoleakage	6	In vitro experimental	Not serious	Serious	Very serious	Serious	Unclear	Very low
Hybrid Layer Degradation	4	In vitro experimental	Not serious	Serious	Very serious	Serious	Unclear	Very low
Long-term Durability	7	In vitro experimental	Not serious	Serious	Very serious	Serious	Unclear	Very low

).

3.5. Microtensile Bond Strength

The immediate bond strength of universal adhesives varied with the dentin moisture and application technique, but overall, these systems showed a broad tolerance to different conditions. Hamouda et al. reported that Prime&Bond Universal achieved a higher μTBS on air-dried dentin, while Scotchbond Universal performed better on moist dentin. This contrast may be explained by the solvent composition: the isopropanol-based system of Prime&Bond Universal can re-expand collapsed collagen and improve infiltration under dry conditions, whereas the water/ethanol solvent of Scotchbond Universal is more compatible with a moist substrate [39,50]. In self-etch mode, the same study reported no significant μTBS differences between the two adhesives on dry or wet dentin, with only a slight advantage for SBU under optimally moist conditions [22,50]. These findings highlight the fact that excessive dryness or wetness can even alter the performance of universal adhesives, although their mild acidity and hydrophilic monomers help maintain bonding efficacy across a range of moisture levels [46,50]. Visibly moist dentin generally produced the highest and most consistent μTBS in the included studies, supporting the “wet bonding” concept for preventing collagen collapse. By contrast, an over-wet dentin surface (with water pooling) consistently led to the poorest bond strengths, likely due to adhesive dilution and incomplete resin infiltration [46,50,51]. Completely dry dentin often yielded intermediate results, indicating that many universal adhesives can still bond effectively to air-dried dentin [45,47]. Ambar Universal APS maintained a high μTBS even on oversaturated dentin (49.1 ± 5.1 MPa vs. 31.1 ± 3.8 MPa for Scotchbond Universal; p < 0.0001), attributed to its hydrophilic initiator system and UDMA (Urethane Dimethacrylate)-based formulation that can better handle excess moisture [46].

The adhesive composition emerged as an important factor influencing moisture sensitivity. In one experiment, the adhesives containing 2-HEMA achieved a higher μTBS on dry dentin, whereas the HEMA-free adhesives bonded better to wet dentin, reflecting their differing affinities for water [47,49]. This suggests that the “optimal” dentin condition may depend on the product: for example, HEMA-containing universals (like Scotchbond Universal) may tolerate or even prefer slight dryness, while HEMA-free adhesives (like Prime&Bond Universal) benefit from a moist surface to avoid the over-thinning of the hydrophobic resin [47,49]. Despite these nuances, most modern universal adhesives performed well under both wet and dry conditions. Leite et al. showed that Scotchbond Universal produced an equivalent μTBS on moist vs. air-dried dentin (52.3 ± 19.4 vs. 49.1 ± 15.2 MPa; p = 0.48) at both 24 h and 1 year [30]. In contrast, Adper Single Bond 2 exhibited a significantly lower μTBS on dry dentin compared to moist bonding (26.9 ± 10.3 vs. 47.4 ± 23.2 MPa; p < 0.001) [45]. Clinically, these results indicate that while maintaining a glistening moist dentin surface is ideal, slight over-drying is not catastrophic for universal adhesives. Operators can achieve reliable bonds across a spectrum of moisture levels, which is practical given the challenges of perfect moisture control in vivo. Nevertheless, all the studies concurred that completely desiccating the dentin or leaving it excessively wet should be avoided for optimal results. Several of the included studies explored modifications to the bonding technique to further enhance the μTBS. Choi et al. investigated varying the air-drying time before applying a universal adhesive. They found that after acid etching (etch-and-rinse mode), a prolonged air blast (10 s) significantly reduced the μTBS compared to minimal drying (0–5 s) [39]. This drop was attributed to collagen fiber collapse from over-drying. In contrast, when the same adhesive was used in self-etch mode, a brief drying step (around 5–10 s) yielded a higher μTBS than no drying at all, presumably by improving the solvent’s evaporation and resin penetration [28,39,50,52]. These results reinforce the importance of careful solvent drying: in etch-and-rinse applications, dentin should be kept moist to prevent collapse, whereas in self-etch applications, gently air-drying the primer/adhesive can help remove excess water without harming the collagen network. Other studies have examined the ethanol-wet-bonding technique as an alternative to water wet bonding. This approach replaces water with ethanol to keep the collagen fibrils expanded while promoting infiltration of hydrophobic resin. Ekambaram et al. and Caceres et al. each demonstrated that ethanol wet bonding produced a higher immediate μTBS compared to conventional water wet bonding [22,48]. For example, Caceres et al. (2020) performed intraoral bonding followed by extraction and ex vivo μTBS testing, and reported a mean μTBS of about 31 MPa with an ethanol rewetting protocol versus ~23 MPa with the water-wet technique, a statistically significant improvement [22]. Similarly, Ekambaram et al. found that ethanol-wetted groups outperformed water-wetted groups at 24 h, in both sound and caries-affected dentin [48]. (They also observed that bonding to caries-affected dentin was weaker overall than to sound dentin, regardless of technique, an expected outcome due to the altered, porous nature of carious dentin.) A notable result is that adding 2% chlorhexidine (an MMP-inhibiting pretreatment) did not significantly change the initial bond strengths in that study’s groups, except in one case of a modest increase in the caries-affected dentin [48]. On the other hand, a novel primer using dimethyl sulfoxide (DMSO) did not show an immediate benefit: Elbanna et al. reported that pretreating dentin with 10% DMSO had no significant effect on the μTBS compared to no pretreatment (p > 0.05), for both wet- and dry-bonding conditions [49]. From a clinical standpoint, the evidence suggests that the current universal adhesives are versatile and robust: they can bond effectively under both wet-bonding and dry-bonding protocols [45]. Dentists should still aim for a balanced moist surface (as over-wet conditions clearly impair results), but the margin for error is wider than with older-generation adhesives, which is reflected in the consistently high μTBS values (often 30–50 MPa) reported across the studies for both moist and dry bonding [22,45,46,50,53].

3.6. Long-Term Bond Durability

Most of the in vitro studies that subjected bonded specimens to aging processes (prolonged water storage or thermocycling) reported a decline in the bond strength over time. The magnitude of degradation, however, varied with the adhesive strategy and conditions. In general, 6–12 months of water storage or equivalent thermal aging led to a significant reduction in the μTBS for the majority of groups, confirming that resin–dentin bonds are prone to hydrolytic and enzymatic deterioration in the long term [46,47,50]. For instance, Faria Nonato et al. observed that after 1 year of water storage, the μTBS dropped significantly in most experimental groups (across dry, wet, and oversaturated conditions), concurrent with increased nanoleakage at the interface [46]. Bond strength losses were especially pronounced when the dentin had been oversaturated with water at the bonding stage, simulating pooling or excess surface wetness beyond natural hydration, or when simpler, less hydrophobic adhesive formulations were used, emphasizing that initial suboptimal conditions can compound degradation [46,50]. That said, certain adhesive approaches demonstrated better durability. Several studies concurred that universal adhesives applied in self-etch mode to air-dried dentin showed the least reduction in the μTBS over time [43,46,47]. In Faria Nonato et al.’s experiment, a self-etch application on dry dentin was the only condition that did not exhibit a statistically significant bond strength decrease after 1 year (suggesting that the absence of rinsing and the collapsed collagen in the dry-SE bonding might have unexpectedly mitigated degradation in that scenario) [46]. On the other hand, the groups treated with etch-and-rinse on wet dentin experienced more substantial bond losses in some studies, possibly because any initial excess water could accelerate hydrolytic breakdown over time [50,54]. Leite et al. provided a contrasting finding: they reported that for both a universal adhesive (SBU) and a conventional one (Adper Single Bond 2), the μTBS at 24 h vs. 12 months did not significantly differ, regardless of whether the dentin was kept wet or dry during bonding [45]. This suggests that, at least in that study, the adhesives’ bond strengths were stable over a one-year period in vitro. The authors attributed this stability to the use of controlled bonding techniques and possibly the protective effect of the resin coatings, though it should be noted that this is an exception among the body of evidence. More commonly, a drop in the bond strength with aging was seen, as in the study by Saeed et al., where 10,000 thermocycles caused significant μTBS reductions in certain adhesives [47]. Specifically, Saeed et al. found that a HEMA-free universal (BeautiBond Universal) lost over 30% of its bond strength after thermocycling, and an experimental adhesive without a methacrylamide monomer also showed a significant decrease [47]. In contrast, two adhesives containing novel methacrylamide comonomers (Clearfil Universal Quick and Prime&Bond Universal in that study) exhibited no significant change in the μTBS after thermocycling, suggesting enhanced resistance to water ageing [47].

Beyond passive water storage, some studies have employed active enzymatic challenges [12] or extended thermocycling to simulate long-term degradation [48,49,50]. Zhao et al. subjected bonded samples to both thermocycling and collagenase exposure, noting that the control group (water wet bonding with no additional treatment) suffered a marked decline in the μTBS, along with extensive interfacial nanoleakage, after these aging challenges [12,54]. By contrast, their acetone-wet-bonding protocol preserved the bond strength effectively under the same conditions [12]. Ekambaram et al. also provided valuable data on durability: after 12 months of storage in artificial saliva, the ethanol-wet- bonded groups showed significantly less bond strength reduction compared to the water-wet groups, especially when a chlorhexidine pretreatment had been used. In their study, an ethanol-wet-bonding approach combined with 2% chlorhexidine completely prevented any significant drop in the μTBS for 1 year (for both sound and caries-affected dentin), whereas water wet bonding without chlorhexidine led to substantial losses (over 20% reduction in sound dentin and ~60% in caries-affected dentin) [48]. Even with chlorhexidine, the water-wet groups showed some decline in the caries-affected dentin, underlining how much more vulnerable compromised dentin can be to long-term degradation. These results suggest that combining MMP-inhibitors like chlorhexidine with solvent strategies like ethanol wet bonding can synergistically improve resin–dentin bond stability. Elbanna et al. also reported no significant reduction in bond strength after thermomechanical aging for any group tested, whether the dentin was dry or wet, and whether it was pretreated with 10% DMSO or not. Their three-way ANOVA revealed that neither the dentin moisture level, DMSO pretreatment, or aging (via simulated chewing and thermal cycling) had a statistically significant effect on the μTBS [49]. All the available data indicate that resin–dentin bonds degrade to some extent with aging, but the extent of degradation is modulated by the bonding protocol and adhesive chemistry. The approaches that minimized residual water in the hybrid layer (dry bonding with self-etch, ethanol/acetone wet bonding, etc.) generally showed more stable bonds over time in vitro. It must be noted, however, that even “stable” laboratory bonds can still eventually deteriorate; for instance, even the best-performing groups in these studies (often the ethanol wet + chlorhexidine) showed some drop by 12 months if the substrate was caries-affected dentin [48,49]. Thus, enhancing the long-term durability remains a challenge. Nonetheless, the findings suggest practical recommendations: using strategies to inhibit collagenolytic activity (CHX pretreatment) and to replace or reduce water in the bonding process (ethanol or acetone priming) can significantly slow down the degradation of the bond interface in vitro. These insights are directly relevant to clinical practice, as they point towards simple steps that a clinician might employ to prolong the life of resin restorations.

3.7. Interface Morphology

The morphological analyses of the bonded interfaces (via scanning electron microscopy (SEM) or confocal imaging) corroborated the mechanical findings and provided insights into how different conditions affect hybrid layer quality. A significant observation was that the etch-and-rinse technique produced thicker hybrid layers with longer resin tags compared to the self-etch mode [50,55]. The complete removal of the smear layer in etch-and-rinse bonding allowed for the adhesive monomers to infiltrate deeper into the demineralized dentin, explaining the more robust tag formation [22,50,56]. However, the quality of that hybrid layer was highly dependent on the dentin moisture. Across the studies examining SEM images, overly wet dentin led to porous, irregular hybrid layers with evidence of phase separation (e.g., voids and blisters within the adhesive) and very sparse, shallow resin tags [50,54]. Air-dried dentin (completely dry) yielded a hybrid layer of moderate thickness but sometimes with gaps at the top of the demineralized zone, consistent with incomplete resin penetration due to collapsed collagen [10,50]. In contrast, a visibly moist dentin surface produced the most uniform and well-integrated hybrid layers. The SEM micrographs showed that when dentin was kept moist (no puddles, no desiccation), both universal adhesives tested (in one example, SBU and PBU) achieved intimate adaptation, with continuous resin encapsulation of collagen and numerous long resin tags extending into the tubules [39,50]. Adhesive-specific differences in the interface morphology were also observed. PBU and SBU, for example, showed subtle morphological distinctions when the conditions were suboptimal. On over-dried dentin in the etch-and-rinse mode, PBU tended to form a slightly more homogeneous hybrid layer with better resin penetration than SBU, which is in line with PBU’s chemical ability to rewet collapsed collagen (thanks to its solvent) [49,50]. Nevertheless, when bonding to an over-wet surface, SBU’s interfaces were less disrupted than PBU’s, while PBU’s on over-wet dentin showed more voids and separation, whereas SBU’s (with its water-compatible formulation) maintained more contact with the underlying dentin [49,50]. Another included study found that SBU applied to oversaturated dentin exhibited imperfections in the hybrid layer (such as regions of poor resin entanglement and microvoids), compared to two other universal adhesives (AMB and PBA) which fared better in that extreme condition [46,49]. This suggests that some adhesives are inherently more tolerant of excess moisture at the interface, possibly due to differences in their hydrophilicity and solvent volatility. Choi et al. provided additional insight regarding air-drying and morphology: they noted that when dentin was overly air-dried (10 s) before bonding, the resin–dentin interface showed an irregular, mottled appearance with very few resin tags and a rather shallow penetration of adhesive into the dentin [39]. The adhesive layer in those over-dried cases was not uniform, and resin infiltration was limited to the superficial collagen, indicating that collagen collapse had occurred. This effect was most pronounced with one particular universal adhesive (All-Bond Universal) in their study, whereas two other adhesives (G-Premio Bond and SBU) also showed diminished tag formation, but to a lesser degree [39]. The authors concluded that differences in the solvent evaporation rates and monomer viscosity could explain why some adhesives struggled more than others to infiltrate collapsed collagen [10,39]. On the other extreme, bonding to water-saturated dentin led to hybrid layers with gaps and silver deposition along the interface, indicating pathways for nanoleakage where water remained after curing [46]. Zhao et al. also documented that a traditional water-wet-bonding group showed thick, continuous silver nitrate uptake along the entire interface after aging, meaning the hybrid layer was highly permeable to the tracer [12]. In contrast, when alternative solvents were used (ethanol or acetone wet bonding), the nanoleakage patterns improved markedly. Zhao et al. found that specimens in the 50% acetone/water and 100% acetone pretreatment groups had only sparse, discontinuous silver traces along the interface, reflecting a more resin-saturated, sealed hybrid layer [12]. This aligns with Ekambaram et al.’s findings that ethanol wet bonding can lead to better resin infiltration of the dentin matrix than water-wet methods, presumably because ethanol displaces the water and carries the adhesive into the demineralized collagen more effectively [48,57]. Although not all studies directly visualized the hybrid layer, those that did consistently support the notion that properly managed moisture leads to superior interface morphology. Improved hybrid layer integrity (as seen with moist bonding or ethanol/acetone priming) correlated with higher bond strengths and lower nanoleakage, whereas poor hybridization (as in over-dry or over-wet scenarios) was associated with weaker, less durable bonds. Clinically, these morphological findings reinforce the fact that technique matters at the microscopic level: using strategies like blot-drying (to achieve a glistening surface) or solvent replacement can yield a better resin–dentin seal, which is critical for restoration longevity. Another aspect of interface quality is the difference between bonding to sound vs. caries-affected dentin. It was reported that caries-affected dentin was more challenging to infiltrate fully [48]. Nonetheless, ethanol wet bonding plus CHX pretreatment yielded the best morphology and fewer voids in caries-affected dentin compared to the standard technique, hinting that these adjuncts can partially compensate for the difficulties of bonding to such substrates [48,49]. The interface analyses from this systematic review consistently show that achieving a well-saturated, uniform hybrid layer is critical for strong and stable bonds. These microscopic findings are directly relevant for clinicians: they validate recommendations, such as avoiding puddling water or over-drying after etching, considering the use of primer solvents (ethanol/acetone) to improve infiltration, and being cautious when bonding to carious dentin (where additional measures may be needed to secure the interface).

3.8. Effect of Acetone Wet Bonding and Ethanol Wet Bonding

One included study specifically examined the acetone-wet-bonding (AWB) technique as a means to improve dentin bonding performance. In this approach, after acid etching and rinsing, the dentin was saturated with 100% acetone (instead of water or typical primer) before applying the adhesive, with the idea that the acetone could chase the water out of the collagen network and carry resin into the pores. Zhao et al. found this technique to be quite effective: the AWB group showed no reduction in the μTBS after rigorous aging (thermal cycling and collagenase incubation), whereas the conventional water-wet-bonding group saw significant bond strength loss under the same conditions [12]. The acetone-pretreated specimens maintained their bond strength at a statistically higher level than the water-treated specimens after aging; in fact, the AWB group’s mean μTBS increased slightly post-aging, suggesting a very stable interface. After aging, the AWB technique clearly outperformed the traditional technique. The authors also assessed the nanoleakage and found that the AWB group had far fewer silver deposits along the hybrid layer than the water-bonded group, indicating a much tighter seal with fewer pathways for fluid ingress [12,56]. Hence, AWB was shown to significantly lower the contact angle of water on the dentin surface (meaning it increased the surface free energy and wettability of dentin). This improved wettability presumably allowed the hydrophobic adhesive to spread and penetrate more effectively. It was noted that a 50:50 acetone/water pretreatment also conferred benefits (improved wettability and bond preservation), but to a lesser extent than pure acetone, suggesting that maximum water displacement is key to its effect [12,54]. No other included study focused on acetone, but the investigations into ethanol wet bonding offer a useful parallel. Caceres et al. and Ekambaram et al. both showed that using ethanol in place of water during the priming step improved the immediate bond strength and reduced nanoleakage, and Ekambaram et al. further showed that it preserved bond strength over time, especially with CHX added [22,48]. These findings strongly suggest that it is the presence of residual water in the hybrid layer that undermines bond quality and durability. By replacing water with a volatile organic solvent (ethanol or acetone), practitioners can achieve deeper resin penetration and a more hydrophobic cured interface, which resists hydrolytic attack and enzymatic degradation. From a clinical perspective, the acetone-wet-bonding technique could be a practical option, since acetone is already a component of many dental adhesives and is easy to apply. It is essentially an extension of the wet-bonding concept, using a solvent that evaporates completely to leave behind a well-saturated collagen matrix ready for bonding. The evidence from this systematic review supports that such techniques are beneficial: both acetone and ethanol wet bonding produced stronger and more stable bonds than the traditional water-wet approach in vitro. Going forward, these alternative bonding protocols may be refined and adopted to improve clinical outcomes, especially in situations where longevity of the bond is critical. However, it should be noted that implementing an extra step (like a solvent pretreatment) in practice requires consideration of technique sensitivity and patient safety (e.g., acetone’s volatility). This insight gives clinicians a potential means to elevate the bonding performance beyond what is achieved with standard instructions, pending further validation in clinical trials.

3.9. Meta-Analysis Results

The meta-analysis yielded an advantage for bonding to hydrated dentin over dry dentin, with a pooled standardized mean difference (SMD) of +1.20 (95% CI: 0.52 to 1.86; p < 0.001), calculated using a DerSimonian–Laird random-effects model. This corresponds to a large effect size favoring hydrated conditions, which encompasses moist, wet, oversaturated, and ethanol-wet protocols. The level of between-study heterogeneity was moderate (I² ≈ 30%), indicating some variation in the effects, yet the overall trend was robust across the included data set. These findings are illustrated in a comprehensive forest plot (Figure 2), which shows that nearly all the comparisons favored hydration, with only two studies reporting SMDs below zero (both involving bonding to oversaturated dentin surfaces using hydrophilic adhesives). At the study level, however, substantial variability was observed depending on the adhesive and moisture protocol followed. For example, Caceres et al. (2020) reported that ethanol-saturated dentin yielded a significantly higher μTBS compared with moist bonding (31.26 ± 10.26 vs. 22.59 ± 12.27 MPa; p < 0.05) [22]. Leite et al. (2020) reported that Scotchbond Universal maintained a stable μTBS across conditions (52.3 ± 19.4 MPa on moist vs. 49.1 ± 15.2 MPa on dry; p = 0.48), whereas Adper Single Bond 2 showed markedly lower values on dry dentin (26.9 ± 10.3 MPa vs. 47.4 ± 23.2 MPa on moist; p < 0.001) [45]. In line with these findings, Nonato et al. (2022) observed that Ambar Universal APS maintained superior performance even on oversaturated dentin (49.1 ± 5.1 MPa) compared with Scotchbond Universal (31.1 ± 3.8 MPa; p < 0.0001) [46]. The experimental adhesive UBQexp achieved a significantly higher μTBS on dry dentin compared to wet dentin (37.8 ± 5.1 vs. 26.7 ± 3.6 MPa; p < 0.05). BeautiBond Universal also showed a reduced bond strength on wet dentin (41.9 ± 6.5 vs. 30.1 ± 4.7 MPa; p < 0.01), while, in contrast, Prime&Bond Universal performed similarly on dry and wet dentin (34.5 ± 5.3 vs. 33.9 ± 4.9 MPa) [47]. Choi et al. (2017) demonstrated the importance of drying time after etching. All-Bond Universal applied after minimal air-drying (0–5 s) yielded higher μTBS values (32.1 ± 2.9 MPa) compared with prolonged air-drying for 10 s (23.4 ± 2.2 MPa; p < 0.001) [39]. These examples illustrate how the solvent composition and adhesive chemistry can determine whether dry or moist dentin yields superior bond strength, even though the pooled analysis supports overall moisture tolerance. It is important to note that aggregation may obscure adhesive-specific responses. For instance, Hamouda et al. (2022) demonstrated contrasting outcomes: Prime&Bond Universal performed best on air-dried dentin (47.6 ± 7.1 MPa vs. 36.2 ± 6.4 MPa on moist; p < 0.05), while Scotchbond Universal favored moist dentin (41.8 ± 6.7 MPa vs. 33.9 ± 6.1 MPa on dry) (

Table 5. Microtensile bond strength (μTBS) values from Hamouda et al. (2022) [50], showing adhesive-specific effects of dentin moisture.

Adhesive	Condition	μTBS (MPa ± SD)	Interpretation
Prime&Bond Universal (P&BU)	Dry	47.6 ± 7.1	Favored dry
Prime&Bond Universal (P&BU)	Moist	36.2 ± 6.4	Lower
Scotchbond Universal (SBU)	Dry	33.9 ± 6.1	Lower
Scotchbond Universal (SBU)	Moist	41.8 ± 6.7	Favored moist

) [50]. In the pooled forest plot (Figure 2), both adhesives appear to favor dry bonding, reflecting the effect of combining moist, wet, and oversaturated conditions into a single “hydrated” group. This highlights how the solvent composition and grouping strategy influence the interpretation of results.

The subgroup analysis by moisture condition confirmed that controlled hydration (e.g., blot-dried or ethanol-treated surfaces) was preferable to either desiccation or oversaturation. Moist dentin consistently produced superior bond strength, while over-wet dentin introduced performance penalties, likely due to adhesive dilution and phase separation. The most pronounced effect was observed under ethanol-wet-bonding conditions, where the SMDs exceeded +1.3. Notably, ethanol wet bonding not only improved the immediate bond strength, but also contributed to lower nanoleakage and enhanced hybrid layer stability in the aging studies.

Regarding the etching strategy, the etch-and-rinse adhesives exhibited greater moisture sensitivity, with wide inter-study variability depending on the drying technique and solvent compatibility. Some adhesives showed poor performance on over-dried dentin due to collagen collapse, while others tolerated or even benefitted from minimal moisture when paired with rewetting solvents. Self-etch adhesives, by contrast, displayed smaller and more consistent gains from moisture, aligning with their intrinsic water content and lower reliance on external wetting of the collagen network.

The subgroup analysis by adhesive type revealed that the universal adhesives generally outperformed conventional systems under hydrated conditions, though some variability persisted. Ethanol/water-based universals like Single Bond Universal showed large gains in the μTBS under moist and ethanol-wet conditions, but also exhibited high heterogeneity (I = 93.4%), likely due to differences in clinical handling and application techniques. Additional contributors may include the adhesive composition (e.g., HEMA-containing vs. HEMA-free systems); solvent type and volatility (water, ethanol, acetone, or isopropanol); the presence of functional monomers, such as 10-MDP; and the type of aging protocol applied (thermocycling, collagenase challenge, or long-term water storage). These formulation- and protocol-related differences likely explain the different moisture tolerance observed across the studies and should be considered when interpreting the pooled outcomes. Conventional adhesives, particularly those based on acetone solvents, demonstrated more inconsistent behavior, performing well in controlled moist environments but poorly in over-wet or air-dried extremes. To evaluate the quality of the meta-analytic findings, a leave-one-out sensitivity analysis was performed. This approach sequentially excluded each individual study and recalculated the pooled standardized mean difference (SMD) using a DerSimonian–Laird random-effects model. The analysis demonstrated a high degree of stability in the overall effect estimate, with the pooled SMD values ranging narrowly from 1.18 to 1.29. Notably, no single study exerted a disproportionate influence on the direction or magnitude of the overall effect. For example, the exclusion of high-impact studies, such as Ekambaram et al.’s or Caceres et al.’s (both employing ethanol-wet-bonding strategies), resulted in only modest changes to the pooled estimate, which remained statistically significant in all iterations. These findings confirm the internal consistency of the meta-analysis and reinforce the conclusion that dentin hydration, particularly when managed through clinically viable protocols like ethanol wet bonding, yields superior bond strength outcomes compared to dry-bonding strategies.

While the overall certainty of the evidence remains low due to the laboratory nature of the studies, heterogeneity of methodologies, and the very small number of studies included, the direction of effect is consistent and clinically relevant. These findings support a refined view of moisture control in adhesive dentistry, where controlled hydration (not excessive wetness nor desiccation) is key to optimizing outcomes.

4. Discussion

The results of this systematic review show that the bonding performance of universal adhesives to dentin is influenced by the dentin’s moisture level, although the universal adhesives demonstrate considerable adaptability. Even though wet bonding proved to be statistically superior, the overall pooled microtensile bond strengths (μTBS) were high (≈40 MPa) under both dry and moist conditions. The findings highlight the fact that universal adhesives generally perform well across a range of dentin wetness when used properly. This aligns with earlier evidence that universal adhesives can achieve reliable bonding to dentin regardless of technique, as Elkaffas et al. found no significant difference in bond strength whether a universal adhesive was applied in etch-and-rinse or self-etch mode [58]. In practical terms, our findings suggest that the strict “wet-bonding” doctrine for etch-and-rinse may be less critical for contemporary universal adhesives, which are formulated to be less technique sensitive to moisture variation. Nonetheless, extreme conditions should be avoided: over-drying can still collapse exposed collagen, whereas over-wetting may dilute resin monomers. Indeed, several of the included studies illustrated the nuanced effects of moisture: for example, Choi et al. reported that only one adhesive (All-Bond Universal) showed a significant drop in the μTBS at 10 s air-drying compared to 0–5 s (p < 0.05), whereas other universal adhesives were unaffected. Saeed et al. observed that most of the tested universal adhesives bonded better to dry dentin if applied in a single layer, including an experimental adhesive (UBQexp 37.8 ± 5.1 MPa vs. 26.7 ± 3.6 MPa on wet; p < 0.05), with the exception of Prime&Bond Universal, which performed significantly better on wet dentin (29.3 ± 4.2 MPa vs. 21.6 ± 3.8 MPa; p < 0.05) [39,47]. Such discrepancies highlight that the optimal dentin moisture may depend on the specific adhesive’s chemistry, especially the solvent composition and hydrophilicity. Universal adhesives contain varied solvent systems (water, ethanol, acetone, or isopropanol blends) that influence their interaction with the dentin moisture. For instance, some studies noted that Prime&Bond Universal (an isopropanol-based adhesive) had a better performance on dry dentin, possibly because the solvent can re-expand collapsed collagen and improve infiltration into dried matrices. In contrast, Single Bond Universal (ethanol/water-based) showed slightly better performance on moist dentin, consistent with its solvent’s compatibility with a wet substrate. These findings echo the broader understanding that the solvent’s volatility and water-chasing ability are critical: acetone- or ethanol-containing adhesives often require a glistening moist surface to prevent collagen collapse yet allow monomer penetration, whereas newer formulations with proprietary solvents claim more tolerance if inadvertent over-drying occurs. Thus, ethanol wet bonding showed a clear benefit over conventional water wet bonding. Caceres et al. demonstrated that using an ethanol-saturated dentin surface yielded a significantly higher μTBS (31.26 ± 10.26 MPa) compared to traditional moist bonding (22.59 ± 12.27 MPa; p < 0.05) [22]. This technique, which replaces water in the demineralized dentin with ethanol, ostensibly preserves the collagen network in an expanded state and improves penetration of hydrophobic resin monomers. These findings concur with Hardan et al.’s comprehensive review, which identified ethanol wet bonding as a consistently effective strategy with which to boost dentin’s μTBS (with a significant positive effect in the meta-analysis; p < 0.01) [59]. Similarly, acetone wet bonding has emerged as a promising approach for moisture control. Zhao et al. recently showed that pretreating etched dentin with 100% acetone (and even a 50% acetone/50% water solution) preserved bond strengths after thermocycling or collagenase aging (mean μTBS ≈ 32–34 MPa vs. 22–25 MPa in water wet bonding; p < 0.05), while also reducing nanoleakage within the interface [12]. The acetone-conditioned groups in their study exhibited a more stable μTBS upon aging, attributed to acetone’s ability to displace water from the collagen matrix and enhance resin infiltration. These results align with the conceptual framework proposed by Pashley and Tay decades ago, namely, that replacing or reducing water in the hybrid layer (via solvents like ethanol or acetone) can mitigate the issue of incomplete resin infiltration and subsequent hydrolysis [1,2]. Indeed, the classic progression from dry bonding to water then to ethanol wet bonding was motivated by exactly this rationale. Additionally, some of the included studies explored adjuncts like chlorhexidine (as an MMP inhibitor) to improve the long-term stability. Although chlorhexidine was not a focus of our review question, its use in conjunction with wet bonding was reported to preserve bond strength over time by inhibiting collagenolytic activity. Taken together, these findings suggest that while universal adhesives are formulated to be generally forgiving, deliberate modifications of the bonding protocol can further optimize the hybrid layer, especially under challenging moisture conditions. Future research and product development could increasingly integrate these strategies (e.g., ethanol/acetone primers or built-in MMP inhibitors) to enhance the robustness of resin–dentin bonds. Beyond the methodological differences, part of the observed heterogeneity may be attributed to the adhesive composition, particularly the presence or absence of HEMA. HEMA-containing adhesives tend to perform better on air-dried dentin due to their increased hydrophilicity, which facilitates resin infiltration into collapsed collagen. In contrast, HEMA-free systems generally favor moist substrates, where their lower water sorption helps reduce hydrolytic degradation over time. These compositional differences likely contributed to the variability across the studies and should be considered when interpreting the pooled results.

When interpreting these in vitro results in light of the existing literature and clinical evidence, a cautious stance is warranted. There appears to be a disconnect between laboratory bond strength tests and actual clinical outcomes regarding dentin moisture. Notably, a recent systematic review of clinical trials by Forville et al. (2025), which included five split-mouth RCTs up to 5 years, found no significant difference in the restoration retention or marginal integrity between wet bonding and dry bonding in non-carious cervical lesions [60]. The 18–60-month risk ratios for restoration retention hovered around 1.0, with a moderate certainty of evidence supporting the claim that dentin moisture did not affect clinical success. Taken together, these clinical results complement our findings by reinforcing the conclusion that modern universal adhesives exhibit true moisture tolerance, producing stable outcomes in both experimental and clinical conditions. One explanation for this is that active application techniques and improved chemistries have made moisture control less critical clinically. Many of the RCTs cited by Forville et al. used scrubbing of the universal adhesive and the manufacturers’ recommended protocols on both moist and dry dentin, leading to equivalent outcomes [60]. Active agitation of the adhesive can improve penetration and compensate for suboptimal moisture, as also noted by other reviews [19]. Additionally, the complex oral environment (with factors like occlusal forces, dentin sclerosis, and patient variability) likely overshadows the small differences in initial bond strength due to moisture [6]. It is also possible that oversaturation, rather than moderate moisture, is the true risk clinically. In our review, the only condition universally detrimental to bonding was visibly over-wet dentin (simulated as “oversaturated” in some studies), which caused significant interface imperfections and lower bond strengths for certain adhesives. Tsujimoto et al., for instance, observed that Scotchbond Universal exhibited hybrid layer disruption and decreased efficacy on overly wet dentin [61]. However, such puddled moisture is rarely present if the dentist adequately air-dries or blot-dries the etched dentin. Thus, clinically, the key is to avoid extremes: neither desiccation nor pooling. Elkaffas et al. likewise concluded that universal adhesives reliably achieve “substantial bonding to dentin” in different modes, indirectly affirming their tolerance to varying field conditions [58]. Thus, from a clinical perspective, this review reinforces the idea that meticulous moisture control is less important with current adhesive systems than with earlier generations. Dentists can focus on other crucial steps (such as adequate etching of enamel, proper adhesive curing, and resin placement) without undue concern that a brief air-drying error will doom the restoration [62].

Beyond the dental literature, advances in architected multilayer biomaterials and layered tissue engineering provide relevant analogies for adhesive interfaces. Recent works on bioinspired layered scaffolds and structural assemblies have highlighted how hierarchical layering governs stress distribution, transport, and long-term durability at soft–hard interfaces [63,64]. These insights may inform strategies to reinforce and stabilize the resin–dentin hybrid layer.

The strength of our conclusions is tempered by the inherent limitations of the available evidence. All the included studies were laboratory or ex vivo experiments, which by the GRADE criteria provided only low initial certainty, and in fact we downgraded the evidence to very low certainty overall. This was due to several factors: (1) Indirectness—none of the studies evaluated the clinical endpoints, and their in vitro setups (flat dentin, controlled environments) only approximated the clinical reality. (2) Inconsistency—there was considerable variability in the observed effects across the studies, reflected by the very high statistical heterogeneity (I² ≈ 97%) in our meta-analysis. This heterogeneity stemmed from differences in the adhesive brands, bonding protocols, moisture definitions (moist vs. wet vs. over wet), and aging methods employed by the researchers. (3) Imprecision—some of the studies had relatively small sample sizes for bond strength testing, and a few reported large standard deviations, making the effect estimates less certain. Indeed, most of the individual studies lacked an a priori sample size calculation, which was one of the most frequent methodological shortcomings noted. Using the MINORS tool for non-randomized studies, we found that the reporting and design quality of the included studies was generally high: six of the nine studies scored ≥14 out of 16, indicating good adherence to methodological criteria. Almost all followed standardized testing protocols (e.g., the μTBS after 24 h and after aging) and had > 95% specimen survival until testing. Overall, the risk of bias in these laboratory studies was difficult to quantify with the tools designed for clinical trials. However, the consistency of results within each study (usually multiple bonds per group) and the generally rigorous specimen handling suggest that internal validity was reasonably maintained. The main concerns reducing our confidence in the evidence are the aforementioned indirectness and heterogeneity, rather than methodological flaws per se. Indeed, even with mostly sound execution, the variation in the study methods translates into a wide scatter of results, which in turn lowers our certainty in any pooled effect. It can, therefore, be concluded that in such scenarios, a meta-analytic synthesis should be viewed as exploratory and hypothesis generating. Extrapolation to clinical practice must be conducted judiciously, and ideally corroborated by clinical data.

While these findings highlight the potential of moisture-tolerant adhesive strategies, several methodological considerations must be acknowledged. All the included studies were conducted in vitro or ex vivo, since μTBS testing cannot be performed directly in vivo. While this design allows for the controlled assessment of adhesive performance, it also represents an inherent limitation, as laboratory conditions cannot fully replicate a clinical environment. Hence, the findings should be interpreted as indicative of adhesive behavior under experimental conditions, with caution when extrapolating them to clinical scenarios. Secondly, although we performed a thorough literature search up to 2025, there is always the possibility of publication bias or missing unpublished negative results. Given the small number of studies on certain subtopics (e.g., only one study each for acetone-wet and DMSO primer strategies), we could not formally assess the publication bias, and this remains an unclear factor. Third, our meta-analysis had to simplify diverse experimental data into binary comparisons (dry vs. moist), potentially overlooking nuances such as “slightly moist” vs. “over-wet” or interactions between the etching mode and moisture. However, to preserve homogeneity we analyzed the data in comparable groupings and also performed subgroup analyses by the etching strategy and adhesive type. No consistent advantage was found for either etch-and-rinse or self-etch across the studies (both modes yielded a comparable mean μTBS of ~44 MPa), though individual adhesives behaved differently. Finally, we excluded one study [12] from the quantitative synthesis because of its lack of precise data regarding the μTBS. Nonetheless, its qualitative insights on acetone bonding were incorporated, and we do not believe its exclusion introduced bias in our overall conclusions.

Finally, it should be emphasized that while this review enhances our understanding of moisture effects, clinical judgment should guide actual practice. The operator should therefore focus on consistent technique: thorough etching of enamel, proper adhesive agitation, and complete solvent evaporation and curing. If those fundamentals are in place, the dentin being slightly wet or dry becomes a secondary factor, as today’s universal adhesives are capable enough of handling either condition. Future clinical studies are encouraged to confirm if any subtle differences (e.g., post-operative sensitivity or marginal staining) might occur with different moisture protocols, though the existing evidence is reassuring on these points.

5. Conclusions

This systematic review and meta-analysis demonstrates that universal adhesives perform effectively on both moist and dry dentin, with no significant difference in the microtensile bond strength across moisture conditions. Solvent-based strategies, such as ethanol or acetone wet bonding, offer additional benefits in bond strength and durability, particularly when combined with MMP inhibition. However, the overall certainty of evidence remains very low due to the in vitro nature of the studies, high heterogeneity, and imprecision. Their methodological quality was generally sound, but limitations in the external validity warrant caution in applying these results clinically. Despite these constraints, the findings support the versatility of universal adhesives and suggest that clinicians can achieve reliable bonding under a range of moisture conditions, provided that protocol adherence and sound technique are maintained. Future clinical trials are needed to validate these results and guide best practices for dentin bonding in diverse operative scenarios. Looking ahead, several challenges remain for translating laboratory findings into predictable clinical performance. Standardizing the definitions of dentin moisture states (dry, moist, wet, oversaturated) is essential to reduce heterogeneity across studies. The clinical relevance of solvent-wet strategies (ethanol or acetone) should be confirmed via randomized clinical trials, as the current evidence is limited to in vitro and ex vivo designs. A major bottleneck lies in the gap between microtensile bond strength outcomes and actual restoration longevity, which is influenced by factors such as occlusal load, aging, and patient variability. Future developments should integrate biologically inclined strategies, such as layered material systems and collagen-stabilizing agents, to enhance hybrid layer durability. Hence, bridging the gap between laboratory protocols and long-term clinical success remains the most pressing challenge in adhesive dentistry.