Computer-aided design and computer-aided manufacturing (CAD-CAM) techniques have become widely popular for prosthetic restorations, because of the streamlined, high-quality results, and the significant improvements in patient comfort [1], [2], [3].

Both resin composites and glass-ceramics are used for such indirect aesthetic restorations [4]. Although glass-ceramic materials are recognized for their aesthetic, mechanical and biocompatibility properties [4], [5], [6], they have some drawbacks, such as brittleness and wear of the opposing teeth [7], [8], [9]. Conversely, resin composites demonstrate a higher propensity for failure (debonding or catastrophic fracture), elevated levels of discoloration, and increased wear [5], [7], [10], [11]. However, when compared to ceramics, resin composites exhibit reduced wear on opposing teeth. [7], [11].

Over the past decade, new CAD-CAM materials have been introduced to the dental market, to combine the advantages of both resin composites and ceramics [8], [12], [13]. Among these materials, dispersed particle-filled composite resins demonstrate enhanced fatigue resistance in ultra-thin occlusal veneers relative to leucite and lithium disilicate ceramics [14] and induce less wear on the opposing teeth [15]. Nonetheless, these biomaterials still exhibit a reduced wear resistance [6], [8], [15] and a slightly higher discoloration rate than ceramics [6], [15], yet a higher translucency [15]. These materials also have lower costs [16], easier intraoral repair[17], and simplified processing requirements [4] compared to ceramics. These advantages make dispersed-filled composite resins a favorable option in specific clinical scenarios, where antagonist wear and cost-efficiency are key considerations.

Clinically, the dispersed particle-filled composite resins show survival rates ranging from 89.0 % to 97.9 % for a two-to-five-year observation period [16]. However, studies emphasize that the success rate of the dispersed particle-filled composite resin restorations (55.6 %) is markedly inferior to that of ceramic CAD/CAM restorations (81.2 %), with frequent observations of restoration debonding [16], [18]. One potential explanation for these failures by debonding is that the resin matrix of CAD/CAM material exhibits an elevated degree of monomer conversion [13], [16]. Thus, the scarcity of monomers available to interact chemically with the luting agent significantly affects the bonding strength of these materials, leading to restoration debonding [13], [16].

Accordingly, surface conditioning prior to luting is crucial for these materials [13], [19], [20]. In recent years, numerous surface treatments have been proposed [17], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], including hydrofluoric acid etching [23], [24], [25], sandblasting with alumina particles (Al2O3) [21], [23], [24], [25], [28], [29], tribochemical silica coating [29], [30], [31], [32], [33], silanization [28], [30], [31] and universal adhesive/primer application [22], [28], [34]. However, the chemical composition of recently introduced dispersed particle-filled composite resins is heterogeneous, [13] and no consensus for their bonding protocol has been reached yet [17], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34].

Among the proposed surface treatments, sandblasting with Al2O3 appears to be efficient [19], [22], [23], [24], [28]. Data show that prior to any chemical treatment, the biomaterial surface must undergo a physical surface treatment to partially remove the resin matrix and expose the filler particles, making them reachable for the chemical treatment [27], [35]. As sandblasting with Al2O3 results in an elevated surface roughness, the surface energy of the indirect restorations is enhanced [25], [35]. Thereby, this treatment yields a superior mechanical attachment of the luting agent to the material [25], [35]. Moreover, two meta-analyses have recommended sandblasting as a superior alternative to hydrofluoric acid etching [17], [20].

But the quality of sandblasting is affected by the particle size and the applied pressure [36]. For instance, particle size has shown a great influence on the bond strength of polycrystalline ceramics [36]. For the dispersed particle-filled composite resins, a range of micro-alumina particle sizes have been proposed for their surface treatment, including 27 µm [23], [24], 30 µm [29], 70 µm [22], with 50 µm being the most frequently used particle size [21], [25], [27]. To the best of our knowledge, the effect of Al2O3 particle size on particle-filled composite resins’ topography has not been studied, and no particles exceeding 100 µm were tested to sandblast these materials. As to the sandblasting pressure, few recommendations exist for dispersed particle-filled composite resins, with one previous work indicating that the sandblasting pressure should not exceed 0.2 MPa to prevent the formation of microcracks on the surface of dispersed particle-filled composite resins [27].

Regarding the chemical surface treatments, when employed as a standalone technique, silane applied onto nanoparticle-filled composite resins displays inferior adhesion properties compared to sandblasting [28]. Hence, the use of both silane and sandblasting are recommended [17], [29], [30], [31], [33] with two meta-analyses indicating that silanization is the most efficient approach when combined with sandblasting [19], [20]. In addition, there is data to support the use of a universal adhesive containing 10-Methacryloyloxydecyl dihydrogen phosphate (MDP) [22], [25], [28].

As to the bonding procedure on the tooth side, in the context of indirect restorations, bonding to the dentin is often involved. Nonetheless, adhesion to dentin poses a formidable challenge due to its tubular structure and organic composition [37], [38]. To streamline these bonding procedures, universal dental adhesives were developed, aiming to reduce the number of steps involved and sometimes the number of bottles of products required [37], [39], [40]. In contrast, two-bottle adhesive systems that utilize a universal adhesive-derived primer have demonstrated superior bond durability, under various aging conditions, on both enamel and dentin, in both self-etch and etch-and-rinse modes, compared to conventional systems [41], [42], [43], [44], [45]. Furthermore, the cured adhesive layer of these two-bottle universal adhesives exhibits higher hydrophobicity [42]. Recent in vitro studies showed similar dentin bonding effectiveness of two-bottle adhesive systems to that of the universal adhesives in self-etch mode [45], [46] and similar [45] of higher [46] bonding effectiveness in etch-and-rinse mode. The etch-and-rinse application mode has shown the best results on micro-tensile bond strength when bonding on air-abraded dentin [37]. As a consequence, it is reasonable to presume that these adhesive systems may offer durable clinical performance and find a widespread use in clinical practice.

The objective of the present in vitro study was to examine the bonding strategies of a dispersed nanoparticle-filled composite resin CAD-CAM material (Cerasmart 270, GC Corp., Japan) to human dentin. The study evaluated the effect of two sandblasting pretreatments with Al2O3 particles on the material side, 50 µm (micro-sandblasting) and 250 µm (macro-sandblasting), as well as the effect of two different adhesive system, G2-Bond Universal (G2, two-bottle adhesive system using a universal adhesive derived primer) and G Premio-Bond (GP, one-bottle universal adhesive system) (GC Corp., Japan) on the dentin side. The research hypotheses were that there would be no difference in the bond strength of Cerasmart 270 between macro-sandblasting and micro-sandblasting, and between G2 and GP universal primers.