For the tooth preparation, a typodont upper right central incisor (Dentoform M-860; Columbia Dentoform Corp) was used [12, 17, 26]. A standard veneer preparation was carried out using the Ceramic Veneer System (CVS) bur set (CVS for porcelain veneers, Komet, Germany). After making depth orientation grooves, the veneer preparation was finalized with a tapered diamond point. The cervical third of the tooth’s facial surface was reduced by 0.4 mm, ending in a chamfer finish line. The preparation was completed 0.5 mm above the cemento-enamel junction. In the middle and occlusal thirds, a 0.6 mm reduction was performed, featuring an incisal butt joint configuration to guarantee correct placement of the MZVs [12]. The tooth was prepared interproximally while preserving the contact areas.

The entire set of experimental procedures was performed by a single, calibrated operator.

For a single prepared tooth, sixty impressions were taken, thirty utilizing traditional impression methods and thirty employing digital impression techniques. The thirty conventional impressions were made with polyvinyl siloxane material (Elite HD plus, Zhermack, Badia Polesine, Italy) and cast using New Plastone II stone (GC Corp, Tokyo, Japan) on a vibrating machine. These final thirty stone casts were scanned using a laboratory scanner (E1 scanner, 3Shape, Copenhagen, Denmark), and a Standard Tessellation Language (STL) file was generated for each cast. All materials were applied following the manufacturer’s guidelines.

Trios 3 Pod intraoral scanner (3Shape, Copenhagen, Denmark) was used to captured digital scan. In total, 60 STL files were collected (30 files from conventional impressions using a laboratory scanner and 30 files from digital impressions using the intraoral scanner). The size of the sample was established based on earlier studies [12, 17, 18].

MZVs were designed using dental software (Dental System 2019, 3Shape, Copenhagen, Denmark), with a cement space setting of 50 μm, and milled in one piece from multi-layered zirconia disks (Ceramic zolid fx multilayer, Amann Girrbach, Austria) using a five-axis milling machine (Ceramill® Motion 2, Amann Girrbach AG, Germany). The undercut areas and sharp edges of each MZV were smoothed and polished with a low-speed fine diamond bur. All MZVs underwent sintering and polishing as per the manufacturer’s guidelines.

The MZVs were carefully positioned on the prepared tooth to assess stability. A thin layer of light-body silicone (Elite HD plus, Zhermack, Badia Polesine, Italy) was spread on both the prepared tooth and the inner surface of the MZVs before fitting the MZVs onto the tooth. An orthodontic force gauge (DTC Orthodontics, Hangzhou, China) was positioned vertically perpendicular to the outer surface of the veneer, with the pushing tip of the gauge resting on the veneer. The veneer was pressed onto the typodont with a consistent force of 3.5 N (equivalent to 350 g) and maintained this pressure, ensuring the force indicator on the gauge remained at 350 g, until the silicone fully polymerized. The veneer was then removed from the typodont using a dental explorer without disturbing the light-body silicone layer adhered to the prepared tooth. A sufficient amount of heavy-body silicone (Elite HD plus, Zhermack, Badia Polesine, Italy) was manually kneaded, rolled into a ball, and placed onto the light silicone layer directly in the middle of the prepared tooth. A glass plate was plated to press the heavy-body silicon onto the prepared tooth, ensuring the edge of the glass plate was parallel to the biting edge of the prepared tooth (Fig. 1a). After the heavy-body silicone had fully solidified and adhered to the light-body silicone layer, the silicone replica was removed, placed on a flat foam piece, and secured with sharp pins. Since the color of the heavy-body silicone differed from the light-body silicone, two contrasting layers was formed (Fig. 1b). The above steps were repeated twice to obtain two silicone replicas: one was sliced in the occlusal-gingival direction, while the other was sliced in the mesial-distal direction.

Illustration of the silicone replica procedure. (A), A glass plate was used to press the heavy silicone onto the prepared tooth, ensuring that the edge of the plate remained aligned to the biting edge of the prepared tooth. (B), Once the heavy-body silicone had completely solidified and bonded with the light body silicone layer, the silicone replica was removed, positioned on a flat piece of foam, and secured with sharp pins

The silicone replicas were sectioned using a scalpel. The cutting lines and measurement points were illustrated in Fig. 2. Line d1: The first cutting line in the occlusal-gingival direction, passing through the highest point of the cervical finish line and perpendicular to the incisal edge; Line d2: parallel to d1 and 2 mm mesial to it; Line d3: parallel to d1 and 2 mm distal to it; Point A1: The highest point of the cervical finish line; Point A2: The intersection of line d1 with the incisal edge; Point A3: the midpoint of line segment A1–A2; Point A4: The midpoint of line segment A1–A3; Point A5: the midpoint of line segment A3–A2; Point B1: the intersection of d2 with the mesial cervical finish line; Point B2: the intersection of d2 with the incisal edge; Point B3: the midpoint of line segment B1–B2; Point B4: the midpoint of line segment B1–B3; Point B5: The midpoint of line segment B3–B2. Similarly, points C1 to C5 are determined on cutting line d3; Line d4: The first cutting line in the buccal-lingual direction, passing through A3 and parallel to the incisal edge; Line d5: Parallel to d4 and 2 mm toward the cervical direction; Line d6: Parallel to d4 and 2 mm toward the incisal edge; Point A6: The intersection of d4 with the mesial finish line; Point A7: The intersection of d4 with the distal finish line; Point A8: The midpoint of line segment A6–A7; Point A9: The midpoint of line segment A6–A8; Point A10: The midpoint of line segment A8–A7; Point B6: The intersection of d5 with the mesial finish line; Point B7: The intersection of d5 with the distal finish line; Point B8: The midpoint of line segment B6–B7; Point B9: The midpoint of line segment B6–B8; Point B10: The midpoint of line segment B8–B7. Similarly, points C6 to C10 are determined on cutting line d6.

Method for determining cutting lines and measurement positions

Subsequently, the silicone replicas were observed under a stereomicroscope (Olympus, SZX2-TR30, Tokyo, Japan) at 20x magnification, integrated with Infinity 1 digital camera (Lumenera, Ontario, Canada). Figure 3 illustrates the cross-section of the silicone replica along with the measurement locations. Measurements were taken using ImageJ software. The marginal gap was measured at 12 positions, including A1, B1, C1, A2, B2, C2, A6, B6, C6, A7, B7, C7, and the internal gap was measured at the remaining 18 positions.

Microscopic images of the silicone replica. (A), occlusal-gingival direction, (B), mesial-distal section. (orange: light-body silicone; purple: heavy-body silicone)

The 12 marginal gap measurement positions were classified into four variable groups: (1) the marginal gap at the cervical area, (2) the marginal gap at the incisal area, (3) the marginal gap at the mesial area, and (4) the marginal gap at the distal area. The 18 internal gap measurement positions were classified into three variable groups: (1) the internal gap at the cervical third, (2) the internal gap at the middle third (vertically), and (3) the internal gap at the incisal third. SPSS 23.0 (SPSS, Chicago, IL) was used to analyze the data. The normality of the data distribution was evaluated using the Shapiro-Wilk test. Statistical differences between the groups were examined using independent T-test, with p-values below 0.05 considered statistically significant.