Effect of build orientation on fracture load and surface roughness of three-unit provisional fixed partial dentures fabricated by digital light processing Authors: Mahnaz Arshad, Mahya Hasanzade, Dimah Haydar Al-Bahrani, Seyed Ali Mosaddad Published: 2026 ABSTRACT ObjectiveTo evaluate the effect of build orientation on the fracture load and surface roughness of three-unit provisional fixed partial dentures (FPDs) fabricated by digital light processing (DLP).MethodsIn this in vitro study, standardized three-unit FPDs were designed and fabricated using a DLP 3D printer at three buccolingual angulations (150°, 180°, and 210°) and three mesiodistal angulations (0°, 15°, and 45°) (n = 12 per group). Surface roughness (Ra, µm) was measured using a profilometer, and fracture load (N) was determined using a universal testing machine. Surface roughness data were analyzed using three-way repeated-measures ANOVA, while fracture load data were analyzed using two-way ANOVA, followed by Tukey post hoc tests (α = 0.05).ResultsBuccolingual and mesiodistal angulations significantly affected surface roughness (p < 0.001) and fracture load (p < 0.001). The lowest surface roughness was observed at 210° buccolingual and 45° mesiodistal angulation, whereas the highest fracture load was recorded at 210° buccolingual and 0° mesiodistal angulation. The effect of location on surface roughness was not significant (p = 0.309).ConclusionsBuild orientation significantly influences the surface and mechanical performance of 3D-printed provisional FPDs. The 210-degree buccolingual angulation showed favorable outcomes under specific mesiodistal conditions; however, the optimal mesiodistal angulation depended on the desired outcome, with 45 degrees improving surface smoothness and 0 degrees enhancing fracture resistance. Introduction Provisional restorations play an important role in fixed partial denture (FPD) treatment by protecting the pulp, preserving periodontal health, and preventing tooth movement [1, 2]. To perform these functions successfully, they should exhibit adequate mechanical and biological properties together with acceptable esthetics [3, 4]. Provisional restorations, commonly fabricated from polymethyl methacrylate (PMMA) or resin composite materials, can be produced using either conventional or digital techniques [5]. Although conventional methods are widely used due to their simplicity and low cost, they present several limitations, including high technique sensitivity, operator-dependent variability, increased chairside time, and reduced reproducibility [5]. In addition, these techniques may result in inferior marginal adaptation, suboptimal surface quality, and inconsistent mechanical performance. Consequently, computer-aided design and computer-aided manufacturing (CAD/CAM) technologies have been increasingly adopted, as they enable greater standardization, improved precision, enhanced mechanical and esthetic properties, and reduced fabrication time compared with conventional approaches [6, 7]. Among digital approaches, additive manufacturing or three-dimensional (3D) printing has become increasingly popular because it enables the fabrication of complex restorations with high detail and low material waste [8]. It is now commonly used for the fabrication of provisional and definitive FPDs within a fully digital workflow [9]. Digital light processing (DLP) is one of the most widely used 3D-printing technologies in dentistry. It relies on layer-by-layer polymerization of liquid photopolymer resins through controlled ultraviolet light exposure, providing high accuracy and precision [10]. In this process, ultraviolet light irradiates acrylic- or epoxy-based light-curing resin, initiating photopolymerization [11]. This technology can produce restorations with fine details and relatively smooth surfaces, making it suitable for dental applications [12, 13]. Surface roughness is a critical parameter that reflects surface irregularity and influences the interaction of materials with light, molecules, and microorganisms, thereby affecting their clinical performance [14]. In dentistry, increased surface roughness promotes plaque accumulation, bacterial adhesion, gingival inflammation, discoloration, and wear of the opposing dentition, while also compromising esthetics [15, 16]. Surface roughness is commonly quantified using the Ra parameter [17, 18]. Fracture load is another clinically relevant property, reflecting the maximum load a restoration can withstand before failure under functional loading; inadequate load-bearing capacity may lead to restoration failure [19, 20]. The properties of 3D-printed restorations are influenced by several manufacturing parameters, particularly build orientation, layer thickness, and post-processing procedures [21, 22]. In 3D printing, the CAD design is converted into a standard tessellation language (STL) file and positioned on the build platform at a selected angle. Appropriate build orientation can improve dimensional accuracy, reduce fabrication time and material consumption, and minimize the need for support structures. Previous studies have demonstrated that printing angulation affects both surface roughness and load-bearing performance, with certain angulations improving surface quality and others enhancing mechanical performance [23–29]. These effects are attributed to variations in layer orientation, interlayer bonding, and polymerization shrinkage, which directly influence the mechanical behavior of printed restorations [30, 31]. However, despite the growing body of evidence on the influence of printing parameters, limited data are available regarding the combined effect of buccolingual and mesiodistal build orientations on both surface roughness and fracture resistance in multi-unit provisional FPDs. Most existing studies have focused on single crowns or evaluated these properties independently [32–35], leaving a gap in understanding the optimal orientation for simultaneously improving surface quality and mechanical performance in 3-unit FPDs [36]. Therefore, the purpose of this in vitro study was to evaluate the effect of build orientation on the surface roughness and fracture load of 3D-printed provisional 3-unit FPDs and to identify the angulation associated with the most favorable performance. The first null hypothesis was that build orientation would have no significant effect on the surface roughness of 3D-printed provisional 3-unit FPDs. The second null hypothesis was that build orientation would have no significant effect on the fracture load of 3D-printed provisional 3-unit FPDs. Materials and methods This in vitro experimental study evaluated provisional 3-unit FPDs fabricated at different printing angulations (Fig. 1). The study protocol was approved by the ethics committee of Tehran University of Medical Sciences (IR.TUMS.DENTISTRY.REC.1402.077). Fig. 1Schematic representation of the study workflow Schematic representation of the study workflow Sample size calculation The minimum sample size for surface roughness assessment was calculated as 11 specimens per printing-orientation group based on a previous study [37], using the one-way ANOVA power analysis module of statistical software (PASS 11; NCSS LLC, Kaysville, UT, USA), with α = 0.05, β = 0.20, a pooled standard deviation of 0.99 μm for Ra, and an effect size of 0.577. For fracture-load assessment, the sample size was recalculated based on previous mechanical-property data [25], using the same statistical assumptions, with α = 0.05, β = 0.20, and an effect size of 0.566. Considering the nine printing-orientation combinations, the minimum calculated sample size for fracture-load assessment was 7 specimens per group. Therefore, 12 specimens were included in each printing-orientation group to satisfy the larger calculated requirement, compensate for possible specimen loss, and maintain a balanced factorial design for both outcomes. Specimen preparation A maxillary second premolar was removed from a dental model (A5AN-500; Nissin Dental Products Inc., Kyoto, Japan) to create the edentulous space. The adjacent maxillary first premolar (#24) and first molar (#26) were prepared as abutments for a 3-unit all-ceramic FPD. Preparation included a 2 mm occlusal reduction, a total convergence angle of 12 degrees, and a 1 mm circumferential deep chamfer finish line, placed 0.5 mm supragingival, using a diamond bur (2-mm torpedo bur; Drendel + Zweiling DIAMANT GmbH, Lemgo, Germany). The prepared teeth and edentulous area were scanned using a desktop dental scanner (Identica T500; Medit Corp., Seoul, Republic of Korea), and the scan data were used to fabricate a cobalt-chromium replica of the master model by casting, providing a standardized, durable reference model for specimen fabrication. Digital design and 3D printing The prepared metal model (Fig. 2) was digitized using a laboratory scanner (UP400 dual 1.3 MP scanner; UP3D Tech Co., Ltd., Shenzhen, China). A 3-unit FPD was designed using CAD software (exocad DentalCAD 3.0; exocad GmbH, Darmstadt, Germany), where the cement space (60 μm), contact diameter (0.8 mm), and contact depth (0.6 mm) were defined (Fig. 3). The finalized design was exported as a standard tessellation language (STL) file and imported into printing software (CHITUBOX; CBD-Tech, Shenzhen, China), where the printing orientation was adjusted, and support structures were automatically generated prior to fabrication. Fig. 2Cobalt–chromium dental model. A, Buccal view. B, Occlusal view Cobalt–chromium dental model. A, Buccal view. B, Occlusal view Fig. 3Computer-aided design (CAD) of the 3-unit fixed partial denture (FPD) in the Exocad software Computer-aided design (CAD) of the 3-unit fixed partial denture (FPD) in the Exocad software The restorations were printed at three buccolingual angulations (150, 180, and 210 degrees) and three mesiodistal angulations (0, 15, and 45 degrees), resulting in nine orientation combinations. The 180-degree buccolingual angulation was obtained by positioning the occlusal surface parallel to the build platform. To obtain 150- and 210-degree buccolingual orientations, the restoration was rotated by + 30 and − 30 degrees around the Z-axis, respectively. For mesiodistal orientation, the 0-degree position was established with the occlusal surface parallel to the platform, followed by rotation by 15 and 45 degrees around the X-axis to obtain the other two mesiodistal angulations. All specimens were fabricated using a digital light processing 3D printer (Saturn 2; ELEGOO, Shenzhen, China) with a build volume of 218 × 123 × 250 mm and a 405-nm ultraviolet light-emitting diode light source. A temporary photopolymer resin containing methacrylic oligomers and photoinitiators was used (Temp Resin; Power Resins, Istanbul, Turkey). Specimens were printed at a layer thickness of 50 μm according to the manufacturer’s instructions [38]. Twelve FPDs were fabricated for each orientation combination (Fig. 4), yielding a total of 108 interim restorations. Fig. 4Representative specimens illustrating the (A) mesiodistal and (B) buccolingual printing angulations used in the study Representative specimens illustrating the (A) mesiodistal and (B) buccolingual printing angulations used in the study Post-processing After printing, the specimens were cleaned by immersion in 99.9% alcohol (Make Take 3D, Tehran, Iran) for 5 min to remove uncured resin. Final polymerization was then performed using a light-curing unit (SUN3 Smart 2 UV/LED lamp; SUNUV, Shenzhen, China) for 45 min [23]. Surface roughness measurement Surface roughness was measured using a surface roughness tester (TR200; TIME Group Inc., Beijing, China). Measurements were performed on the buccal surface of each 3-unit FPD by moving the stylus from the occlusal toward the cervical region across the premolar retainer, pontic, and molar retainer. Three measurements were obtained for each location, and the Ra values were recorded. The mean Ra value of the three readings at each location was calculated, and these values were used for statistical analysis. Fracture load measurement Fracture load was evaluated using a universal testing machine (Z050; ZwickRoell GmbH & Co. KG, Ulm, Germany). Each FPD specimen was seated on the prepared abutments of the cobalt–chromium master model, with the pontic positioned centrally beneath the loading tip. The specimen was carefully aligned so that a 4-mm-diameter spherical loading tip contacted the center of the occlusal surface of the pontic. No cementation or mechanical clamping was used; however, the specimen was stabilized by passive seating on the cobalt–chromium abutments, and a small initial contact was established before loading to prevent displacement. The load was applied vertically at a crosshead speed of 1 mm/min until failure, and the maximum load at failure was recorded in newtons (N). This configuration was used to standardize the testing procedure while maintaining specimen support on the same reference model used for fabrication. All surface roughness measurements and fracture-load tests were performed by a single trained operator using the same measurement protocol to minimize operator-related variability. Statistical analysis Statistical analysis was performed using statistical software (IBM SPSS Statistics, version 20.0; IBM Corp., Armonk, NY, USA) at a significance level of α = 0.05. For surface roughness, the effects of buccolingual angulation, mesiodistal angulation, and measurement location (premolar, pontic, and molar), as well as their interaction effects, were analyzed using a three-way repeated-measures ANOVA, with location considered the within-specimen factor. For fracture load, the effects of buccolingual and mesiodistal angulations and their interaction were analyzed using two-way ANOVA. Post hoc pairwise comparisons were performed using the Tukey test. Homogeneous subsets were generated to identify the orientation combinations associated with the most favorable outcomes. Results Surface roughness Table 1 presents the mean Ra values of specimens printed at different buccolingual and mesiodistal angulations. The highest surface roughness was observed in specimens printed at 150-degree buccolingual and 15-degree mesiodistal angulation, whereas the lowest value was recorded for the 210-degree buccolingual and 45-degree mesiodistal combination. Table 1Mean (SD) surface roughness (Ra, µm) of provisional 3-unit fixed partial dentures printed at different buccolingual and mesiodistal angulationsAngulationBuccolingualMesiodistal150 degrees180 degrees210 degrees0 degrees4.55 (1.57)3.03 (0.78)1.57 (1.00)15 degrees4.56 (1.00)1.91 (0.93)1.83 (0.55)45 degrees2.83 (1.00)1.31 (1.00)1.13 (0.48) Table 1Mean (SD) surface roughness (Ra, µm) of provisional 3-unit fixed partial dentures printed at different buccolingual and mesiodistal angulationsAngulationBuccolingualMesiodistal150 degrees180 degrees210 degrees0 degrees4.55 (1.57)3.03 (0.78)1.57 (1.00)15 degrees4.56 (1.00)1.91 (0.93)1.83 (0.55)45 degrees2.83 (1.00)1.31 (1.00)1.13 (0.48) Mean (SD) surface roughness (Ra, µm) of provisional 3-unit fixed partial dentures printed at different buccolingual and mesiodistal angulations Analysis of surface roughness showed significant effects of buccolingual angulation (p < 0.001), mesiodistal angulation (p < 0.001), and their interaction (p < 0.001). In contrast, the effect of location (premolar, pontic, and molar) was not significant (p = 0.309). Likewise, the interactions between location and buccolingual angulation (p = 0.072), location and mesiodistal angulation (p = 0.163), and the three-way interaction among location, buccolingual angulation, and mesiodistal angulation (p = 0.305) were not significant. Pairwise comparisons showed that, at 180-degree buccolingual angulation, the 45-degree mesiodistal angulation produced significantly lower surface roughness than the 0-degree and 15-degree mesiodistal angulations (p < 0.05). At 210-degree buccolingual angulation, the 45-degree mesiodistal angulation yielded the lowest surface roughness overall, whereas the 0-degree mesiodistal angulation showed the second lowest values. In contrast, the 150-degree buccolingual angulation produced the highest surface roughness values across the mesiodistal groups, particularly when combined with the 15-degree mesiodistal angulation (p < 0.05). Overall, the 210-degree buccolingual and 45-degree mesiodistal combination was associated with the most favorable surface roughness outcome (Table 2; Fig. 5). Table 2Pairwise comparisons of surface roughness among printing angulations (Tukey post hoc test)Comparison setComparisonp valueMesiodistal angulation at 150-degree buccolingual angle0 vs. 151.0000 vs. 45< 0.001*15 vs. 45< 0.001*Mesiodistal angulation at 180-degree buccolingual angle0 vs. 15< 0.001*0 vs. 45< 0.001*15 vs. 450.021*Mesiodistal angulation at 210-degree buccolingual angle0 vs. 150.2860 vs. 450.037*15 vs. 45< 0.001*Buccolingual angulation at 0-degree mesiodistal angle150 vs. 180< 0.001*150 vs. 210< 0.001*180 vs. 210< 0.001*Buccolingual angulation at 15-degree mesiodistal angle150 vs. 180< 0.001*150 vs. 210< 0.001*180 vs. 2100.921Buccolingual angulation at 45-degree mesiodistal angle150 vs. 180< 0.001*150 vs. 210< 0.001*180 vs. 2100.673*Statistically significant at p < 0.05 Table 2Pairwise comparisons of surface roughness among printing angulations (Tukey post hoc test)Comparison setComparisonp valueMesiodistal angulation at 150-degree buccolingual angle0 vs. 151.0000 vs. 45< 0.001*15 vs. 45< 0.001*Mesiodistal angulation at 180-degree buccolingual angle0 vs. 15< 0.001*0 vs. 45< 0.001*15 vs. 450.021*Mesiodistal angulation at 210-degree buccolingual angle0 vs. 150.2860 vs. 450.037*15 vs. 45< 0.001*Buccolingual angulation at 0-degree mesiodistal angle150 vs. 180< 0.001*150 vs. 210< 0.001*180 vs. 210< 0.001*Buccolingual angulation at 15-degree mesiodistal angle150 vs. 180< 0.001*150 vs. 210< 0.001*180 vs. 2100.921Buccolingual angulation at 45-degree mesiodistal angle150 vs. 180< 0.001*150 vs. 210< 0.001*180 vs. 2100.673*Statistically significant at p < 0.05 Pairwise comparisons of surface roughness among printing angulations (Tukey post hoc test) *Statistically significant at p < 0.05 Fig. 5Error bar plot of surface roughness (Ra) values of specimens printed at different mesiodistal and buccolingual angulations. BL, buccolingual; MD, mesiodistal Error bar plot of surface roughness (Ra) values of specimens printed at different mesiodistal and buccolingual angulations. BL, buccolingual; MD, mesiodistal Fracture load Table 3 presents the mean fracture load values of specimens printed at different buccolingual and mesiodistal angulations. Buccolingual angulation had a significant effect on fracture load (p < 0.001), and the interaction between buccolingual and mesiodistal angulations was also significant (p < 0.001). However, mesiodistal angulation alone did not significantly affect fracture load (p = 0.266). The highest fracture load was recorded for specimens printed at 210-degree buccolingual and 0-degree mesiodistal angulation, whereas the lowest value was observed for the 150-degree buccolingual and 0-degree mesiodistal combination. Table 3Mean (SD) fracture load values (N) of provisional 3-unit fixed partial dentures printed at different buccolingual and mesiodistal angulationsAngulationBuccolingualMesiodistal150 degrees180 degrees210 degrees0 degrees1326.55 (398)1827.59 (126)1903.70 (169)15 degrees1782.25 (124)1852.74 (152)1716.58 (153)45 degrees1664.59 (486)1766.45 (225)1753.17 (174) Table 3Mean (SD) fracture load values (N) of provisional 3-unit fixed partial dentures printed at different buccolingual and mesiodistal angulationsAngulationBuccolingualMesiodistal150 degrees180 degrees210 degrees0 degrees1326.55 (398)1827.59 (126)1903.70 (169)15 degrees1782.25 (124)1852.74 (152)1716.58 (153)45 degrees1664.59 (486)1766.45 (225)1753.17 (174) Mean (SD) fracture load values (N) of provisional 3-unit fixed partial dentures printed at different buccolingual and mesiodistal angulations Post hoc Tukey comparisons demonstrated that, at 0-degree mesiodistal angulation, specimens printed at 180-degree and 210-degree buccolingual angulations had significantly higher fracture load than those printed at 150 degrees (p < 0.001), with no significant difference between the 180-degree and 210-degree groups (p = 0.756). In addition, within the 150-degree buccolingual group, the 15-degree mesiodistal angulation showed significantly higher fracture load than the 0-degree mesiodistal angulation (p = 0.013), whereas within the 210-degree buccolingual group, the 0-degree mesiodistal angulation showed significantly higher fracture load than the 15-degree mesiodistal angulation (p = 0.025). Overall, the 210-degree buccolingual and 0-degree mesiodistal combination yielded the highest fracture load, while the 150-degree buccolingual and 0-degree mesiodistal combination yielded the lowest fracture load (Table 4; Fig. 6). Table 4Pairwise comparisons of fracture load among printing angulations (Tukey post hoc test)Comparison setComparisonp valueMesiodistal angulation at 150-degree buccolingual angle0 vs. 150.013*0 vs. 450.07915 vs. 450.718Mesiodistal angulation at 180-degree buccolingual angle0 vs. 150.9330 vs. 450.66715 vs. 450.452Mesiodistal angulation at 210-degree buccolingual angle0 vs. 150.025*0 vs. 450.08015 vs. 450.860Buccolingual angulation at 0-degree mesiodistal angle150 vs. 180< 0.001*150 vs. 210< 0.001*180 vs. 2100.756Buccolingual angulation at 15-degree mesiodistal angle150 vs. 1800.463150 vs. 2100.512180 vs. 2100.068Buccolingual angulation at 45-degree mesiodistal angle150 vs. 1800.726150 vs. 2100.788180 vs. 2100.994*Statistically significant at p < 0.05 Table 4Pairwise comparisons of fracture load among printing angulations (Tukey post hoc test)Comparison setComparisonp valueMesiodistal angulation at 150-degree buccolingual angle0 vs. 150.013*0 vs. 450.07915 vs. 450.718Mesiodistal angulation at 180-degree buccolingual angle0 vs. 150.9330 vs. 450.66715 vs. 450.452Mesiodistal angulation at 210-degree buccolingual angle0 vs. 150.025*0 vs. 450.08015 vs. 450.860Buccolingual angulation at 0-degree mesiodistal angle150 vs. 180< 0.001*150 vs. 210< 0.001*180 vs. 2100.756Buccolingual angulation at 15-degree mesiodistal angle150 vs. 1800.463150 vs. 2100.512180 vs. 2100.068Buccolingual angulation at 45-degree mesiodistal angle150 vs. 1800.726150 vs. 2100.788180 vs. 2100.994*Statistically significant at p < 0.05 Pairwise comparisons of fracture load among printing angulations (Tukey post hoc test) *Statistically significant at p < 0.05 Fig. 6Error bar plot of fracture load (N) of specimens printed at different buccolingual and mesiodistal angulations. BL, buccolingual; MD, mesiodistal Error bar plot of fracture load (N) of specimens printed at different buccolingual and mesiodistal angulations. BL, buccolingual; MD, mesiodistal Discussion This in vitro study evaluated the effect of buccolingual and mesiodistal build orientation on the surface roughness and fracture load of 3D-printed provisional 3-unit FPDs. Build orientation significantly affected both surface roughness and fracture load; therefore, both null hypotheses were rejected. The lowest surface roughness was obtained with the 210-degree buccolingual and 45-degree mesiodistal angulation, whereas the highest fracture load was recorded with the 210-degree buccolingual and 0-degree mesiodistal angulation. These findings indicate that the optimal printing orientation may differ depending on whether surface quality or mechanical performance is the primary objective. The superior surface quality observed at 210-degree buccolingual and 45-degree mesiodistal angulation can be attributed mainly to the effect of layer orientation on surface topography. In layer-by-layer additive manufacturing, the stair-step effect is a major determinant of surface irregularity, particularly on surfaces that are more parallel to the build platform [39]. Tilting the restoration buccolingually to 210 degrees likely reduced the extent of this effect on the buccal surface by altering the relationship between the restoration contour and the printed layers. The additional 45-degree mesiodistal angulation may have further improved the surface profile by distributing the layers more obliquely along the restoration’s curvature, thereby reducing abrupt layer transitions. This interpretation agrees with previous reports showing that angulated printing generally produces lower roughness than flatter orientations [40]. It is also consistent with the findings of Yu et al. [24], who reported favorable surface quality and intaglio trueness within the 150- to 210-degree range, with the lowest root mean square deviation for first molars observed at 210 degrees. Clinically, this finding is relevant because smoother provisional surfaces are less likely to promote plaque accumulation, bacterial adhesion, discoloration, and soft tissue irritation [41, 42]. The present surface roughness findings are generally in line with studies showing that build orientation can influence the topography of printed provisional restorations, although the optimal angulation has not been consistent across investigations. Alharbi et al. [25], Arnold et al. [40], and Sahni et al. [35], all reported that build direction significantly affects the surface characteristics of printed restorations, supporting the present finding that orientation is a relevant determinant of surface texture. Likewise, Alqarawi [31], in a recent systematic review and meta-analysis, concluded that printing orientation affects surface roughness, although the direction of the effect may vary across studies. In contrast, de Castro et al. [29] found that build orientation did not significantly affect roughness-related variables in printed provisional resins. This discrepancy may be explained by major methodological differences, since de Castro et al. [29] evaluated disc-shaped specimens rather than multi-unit prostheses, used different resin systems and printers, and assessed surface behavior after toothbrushing and ultraviolet aging. Compared with flat disc specimens, a 3-unit FPD has a more complex anatomy, greater curvature, pontic contours, and a greater variability in the need for support structures, all of which may amplify the influence of orientation on surface roughness. Therefore, the current findings suggest that orientation-related effects may be more pronounced in anatomically complex provisional prostheses than in simplified standardized specimens. The highest fracture load observed at 210-degree buccolingual and 0-degree mesiodistal angulation can be explained by the relationship between the printed layers and the direction of force application. In additively manufactured restorations, interlayer interfaces are often structural weaknesses because they are more susceptible to incomplete bonding and crack propagation under load [43]. When the restoration is oriented more favorably with respect to the loading direction, stresses may be distributed more efficiently through the layered structure, thereby improving resistance to failure. In the present study, the 210-degree buccolingual orientation likely produced a more favorable layer arrangement under loading at the pontic, whereas the 150-degree buccolingual orientation may have led to a less favorable interlayer configuration, resulting in earlier failure. This explanation is consistent with studies demonstrating better mechanical performance when the printed layer arrangement is more favorable relative to the loading axis [25, 44]. The present mechanical findings are also supported by several recent investigations showing that printing orientation significantly affects the fracture behavior of interim restorations, although the most favorable angulation varies across studies. Diken Turksayar et al. [27] and Sahni et al. [35] found that 0-degree printed 3-unit provisional restorations showed higher fracture load than some alternative orientations, which is partly consistent with the present study because the highest fracture load here was also associated with a 0-degree mesiodistal component. Alkhateeb et al. [33] similarly reported that 0-degree and 45-degree orientations produced higher fracture load than 90 degrees, with 45 degrees showing the highest values overall. Reymus et al. [34] also confirmed that build direction significantly affects the fracture load of printed FDPs. Together, these findings support the present conclusion that orientation significantly influences load-bearing performance. However, direct comparison of the optimal angle remains difficult [45] because the available studies differ substantially in how orientation is defined, whether the long axis was positioned occlusally, buccally, distally, mesiodistally, or buccolingually, and whether the analysis was performed on single crowns, bar specimens, or multi-unit FDPs. Not all studies have identified the same orientation as mechanically favorable. Mudhaffer et al. [28] reported that 90-degree orientation produced the highest flexural properties, and Aljehani et al. [32] also found that 90-degree printed crowns exhibited the highest fracture resistance. Ibrahim et al. [26] likewise reported higher fracture resistance for 0-degree than for 90-degree printed provisional crowns, with 30-degree crowns showing a favorable fit. These differences from the present findings likely reflect variations in specimen geometry, the outcome variable, and the testing configuration. Mudhaffer et al. [28] evaluated bar-shaped specimens according to ISO 4049, which yields a highly standardized stress distribution that differs substantially from that observed when loading a 3-unit FPD at the pontic. Aljehani et al. [32] and Ibrahim et al. [26] investigated single crowns rather than multi-unit prostheses, and single crowns do not reproduce the connector-related stress concentration and span-related deformation that occur in 3-unit FPDs. In addition, the resin formulation, printer system, layer thickness, post-curing protocol, and aging conditions differed among studies. These variables are known to affect polymer conversion, interlayer bonding, and overall mechanical performance, which may explain why different studies identify different optimal orientations [21]. Although the present study did not include conventionally fabricated provisional FPDs, the findings can be interpreted in relation to previous studies comparing conventional and digital provisional restorations. Conventional direct provisional restorations, commonly fabricated from PMMA or bis-acrylic resin materials, remain widely used; however, their surface and mechanical behavior may be influenced by polymerization shrinkage, porosity, finishing and polishing procedures, and operator-dependent variability [1, 3]. Previous studies have shown that digitally fabricated provisional restorations may provide improved standardization and mechanical performance compared with direct handmade provisional restorations [1]. More recent comparative studies evaluating conventional, subtractive, and additive techniques have also indicated that fracture resistance and surface characteristics vary according to the fabrication method and material, with subtractively manufactured provisional restorations often showing favorable mechanical performance, whereas additively manufactured restorations are more dependent on printing and post-processing parameters [4, 5]. Therefore, although the present fracture-load and surface roughness values cannot be directly compared with conventional provisional FPDs because of differences in specimen geometry, material composition, fabrication workflow, and testing configuration, the results support the importance of optimizing DLP printing parameters to improve the surface and mechanical performance of provisional restorations. Layer thickness should also be considered when interpreting the present findings. Although build orientation was the experimental variable in this study, layer thickness may contribute to both surface roughness and fracture behavior of 3D-printed provisional restorations. A reduced layer thickness may decrease the stair-step effect by producing smaller vertical transitions between successive layers, thereby improving surface smoothness and reducing the need for post-processing. In addition, thinner layers may influence polymerization uniformity, interlayer bonding, dimensional adaptation, and the presence of internal defects, which may contribute to improved durability and load-bearing capacity [39]. Previous studies have shown that layer thickness and post-printing conditions can affect the flexural strength, hardness, fit, and fracture resistance of 3D-printed dental resins and provisional crowns [10, 26, 38]. However, because all specimens in the present study were printed at a fixed layer thickness of 50 μm, the isolated effect of layer thickness could not be assessed. Future investigations should evaluate the combined effect of build orientation and reduced layer thickness on the surface and mechanical performance of 3D-printed provisional FPDs. An important observation of the present study is that the orientation associated with the lowest surface roughness was not the same as that associated with the highest fracture load. This suggests that optimization of build orientation should be guided by the primary clinical requirement of the restoration. A 210-degree buccolingual angulation appeared advantageous for both variables overall; however, the mesiodistal orientation determined whether surface smoothness or load-bearing performance was favored. Therefore, the most appropriate printing strategy may depend on whether the clinical priority is improved surface quality or greater resistance to functional loading. This interpretation is consistent with the recent systematic review and meta-analysis by Alqarawi [31], which concluded that printing orientation affects several physical and mechanical properties of provisional restorations, although the favorable direction varies according to the property being investigated. From a clinical standpoint, these findings indicate that build orientation should be selected strategically rather than solely on the basis of printing convenience or production efficiency. When smoother external surfaces are desired to facilitate plaque control and soft tissue health, a 45-degree mesiodistal angulation combined with a 210-degree buccolingual angulation may be preferable. In contrast, when higher resistance to fracture is more critical, such as in provisional restorations expected to withstand greater occlusal demands, a 0-degree mesiodistal angulation combined with a 210-degree buccolingual angulation may be more suitable. At the same time, post-curing conditions, aging, and material selection can also substantially modify fracture load [33, 34], meaning that orientation should be optimized together with these additional manufacturing variables rather than considered in isolation. Careful selection of build orientation may therefore improve both the biologic and mechanical performance of 3D-printed provisional FPDs. This study has some limitations. Only one temporary photopolymer resin and one DLP printer were evaluated, and the scanner, CAD software, slicing software, printer, and resin were obtained from different manufacturers; therefore, the findings may not be directly generalizable to other materials, systems, or closed digital workflows. A conventionally fabricated provisional FPD control group was not included, precluding direct comparison between DLP-fabricated and traditionally fabricated restorations. In addition, no artificial aging was performed, and fracture load was assessed under monotonic static loading rather than fatigue-based cyclic loading; therefore, the influence of thermal, mechanical, and moisture-related degradation on the tested orientations remains unclear, particularly because aging and post-curing protocols can significantly affect the fracture load of printed restorations [28, 33, 34]. Surface roughness was assessed only by linear profilometry on anatomically contoured buccal surfaces, which may have introduced variability related to curvature, stylus trajectory, and local surface irregularities; thus, the Ra values should be interpreted mainly as relative comparisons among printing orientations rather than definitive absolute roughness values. Finally, fracture load was measured using a standardized loading configuration in which the restorations were seated on the prepared abutments of the cobalt–chromium master model without cementation or mechanical clamping. Although this approach enabled direct comparison among printing orientations under controlled conditions, it does not fully reproduce the stress distribution of clinically cemented FPDs supported by prepared abutments. Future studies should evaluate different resin–printer–software combinations, include conventionally fabricated controls, incorporate three-dimensional surface characterization, and test cemented restorations mounted on standardized abutments after thermomechanical aging and cyclic loading. Conclusion Based on the findings of this study, it can be concluded that: Build orientation significantly influences both the surface roughness and fracture load of 3D-printed provisional 3-unit fixed partial dentures.The 210-degree buccolingual angulation showed favorable outcomes under specific mesiodistal conditions, yielding the lowest surface roughness when combined with a 45-degree mesiodistal angulation and the highest fracture load when combined with a 0-degree mesiodistal angulation.The optimal printing orientation depends on the desired clinical outcome, with different angulation combinations favoring either improved surface smoothness or enhanced load-bearing capacity. Build orientation significantly influences both the surface roughness and fracture load of 3D-printed provisional 3-unit fixed partial dentures. Build orientation significantly influences both the surface roughness and fracture load of 3D-printed provisional 3-unit fixed partial dentures. The 210-degree buccolingual angulation showed favorable outcomes under specific mesiodistal conditions, yielding the lowest surface roughness when combined with a 45-degree mesiodistal angulation and the highest fracture load when combined with a 0-degree mesiodistal angulation. The 210-degree buccolingual angulation showed favorable outcomes under specific mesiodistal conditions, yielding the lowest surface roughness when combined with a 45-degree mesiodistal angulation and the highest fracture load when combined with a 0-degree mesiodistal angulation. The optimal printing orientation depends on the desired clinical outcome, with different angulation combinations favoring either improved surface smoothness or enhanced load-bearing capacity. The optimal printing orientation depends on the desired clinical outcome, with different angulation combinations favoring either improved surface smoothness or enhanced load-bearing capacity.