The internal fit and fracture resistance of stainless-steel crowns and porcelain fused to metal crowns in intact and severely damaged permanent molars: an in vitro study Authors: Horieh Moosavi, Fatemeh Khorakian, Fatemeh Sedaghati, Ali Abkar, Hamidreza Baradaran Kashani Published: 2026 ABSTRACT AbstractAimSevere caries in mixed dentition frequently affects mandibular first permanent molars, often necessitating full-coverage restorations. Stainless steel crowns (SSCs) and porcelain-fused-to-metal (PFM) crowns are among the most commonly used options. This study aimed to compare the internal fit and fracture resistance of SSCs and PFMs in both intact and restored mandibular first molars.Materials and methodsForty extracted mandibular first molars were randomly divided into four groups: (1) intact teeth–SSCs, (2) intact teeth–PFMs, (3) restored teeth–SSCs, and (4) restored teeth–PFMs. In restored samples, access cavities and mesiodistal boxes were prepared and restored with composite resin. Following tooth preparation, PFM crowns were fabricated, and appropriate SSCs were selected. The internal fit was assessed using the silicone replica technique under a digital microscope. Crowns were then cemented with glass ionomer cement, and fracture resistance was measured using a universal testing machine. Data were analyzed using two-way ANOVA, independent t-test, and Chi-square test (α = 0.05).ResultsSSCs demonstrated significantly greater marginal gap (intact: 313.20 ± 44.46 μm; restored: 304.90 ± 62.03 μm), axial gap, and total internal gap compared with PFMs (marginal gap: intact: 183.00 ± 41.29 μm; restored: 222.10 ± 38.25 μm; p < 0.001). Tooth condition had no significant effect on SSC adaptation; however, PFM marginal and total internal gap values increased significantly following restoration (p = 0.048). PFMs demonstrated significantly higher fracture resistance than SSCs under both conditions (p < 0.001). Restored SSCs exhibited significantly higher fracture resistance than intact SSCs (p = 0.029), while no significant difference was observed between intact and restored PFMs. SSCs showed a significantly higher proportion of repairable failures (40–80%) compared with PFMs, in which all fractures were non-repairable (p < 0.001).ConclusionsUnder the conditions of this in vitro study, PFMs demonstrated superior internal adaptation and overall fracture resistance compared with SSCs; however, SSCs exhibited more favorable and repairable fracture patterns. Long-term clinical studies are warranted to validate these findings.Keywords: Dental Crowns, Dental Materials, Mechanical Strength, Stainless Steel, Metal-Ceramic Crowns Introduction The first permanent molars are among the most caries-susceptible teeth in the dentition, frequently exhibiting severe destruction during the mixed dentition period, a stage characterized by ongoing jaw development and transitional occlusion. When coronal integrity is substantially compromised, full-coverage restorations become necessary to prevent fracture and restore function [1]. Among the available options, stainless steel crowns (SSCs) and porcelain-fused-to-metal (PFM) crowns represent two fundamentally different restorative approaches, each with distinct mechanical properties, clinical indications, and technical demands. PFM crowns offer aesthetic appeal and precise adaptation; however, their fabrication requires considerable tooth reduction and laboratory precision, which may not be well-suited to the dynamic dentoalveolar environment of growing patients. In mixed dentition, continuing jaw development leads to changes in tooth position, occlusal relationships, and clinical crown height, all of which may compromise the long-term fit and retention of definitively cemented restorations [2]. SSCs, by contrast, are prefabricated, chairside-adapted restorations that can be placed in a single visit with minimal tooth reduction. Their cost-effectiveness, resistance to recurrent caries, and ability to maintain occlusal function and arch integrity have been well-documented, with clinical success rates consistently surpassing those of direct composite restorations [3–5]. These attributes make SSCs particularly advantageous in complex clinical scenarios, including patients with molar-incisor hypomineralization, amelogenesis imperfecta, or dentinogenesis imperfecta, as well as in cases involving partially erupted teeth or uncooperative patients [6]. As such, SSCs are widely regarded as the interim restoration of choice for young permanent molars until skeletal maturity and stable occlusion are achieved, typically around age 18 [7]. Despite their clinical utility, the long-term performance of SSCs is contingent on adequate marginal adaptation. Marginal discrepancies have been associated with microleakage, periodontal complications, pulpal pathology, and recurrent caries, collectively contributing to reported failure rates in the literature [3, 8]. Proper crown selection, along with careful trimming, crimping, and polishing, has been advocated to optimize fit and reduce these risks [9]. While the individual mechanical and adaptive properties of SSCs and PFM crowns have been investigated separately, comparative data examining both internal fit and fracture resistance, particularly in the context of severely damaged permanent molars, remain limited. The present study was therefore designed to address this gap by evaluating and comparing the internal fit and fracture resistance of SSCs and PFM crowns in both intact and extensively restored mandibular first permanent molars. Materials and methods Sample preparation Forty healthy human mandibular first molars, extracted due to periodontal disease, were collected for this study. Inclusion criteria comprised the absence of caries, root fractures, and resorption. Teeth were selected within a standardized dimensional range (mean ± 10%). All samples were used within two months of extraction. Following extraction, teeth were immediately stored in 0.5% thymol solution, then transferred to 0.9% saline at room temperature until the experiment commenced. Each tooth was mounted vertically in self-cure acrylic resin to standardize the positioning during preparation and testing. Samples were randomly allocated to four groups of ten: Group 1: Intact teeth- SSC. Group 1: Intact teeth- SSC. Group 2: Intact teeth - PFM. Group 2: Intact teeth - PFM. Group 3: Restored teeth - SSC. Group 3: Restored teeth - SSC. Group 4: Restored teeth - PFM. Group 4: Restored teeth - PFM. In Groups 3 and 4, standardized access cavities and MOD preparations were created using a high-speed diamond bur, with proximal boxes 4 mm deep and a buccolingual width of 5 mm at the CEJ level (Fig. 1-B). Burs were replaced after every five preparations to ensure consistent cutting efficiency. Sample preparation stages. (A) Intact extracted mandibular first molar prior to preparation. (B) Standardized MOD cavity preparation with proximal boxes in the restored group. (C–F) Sequential steps of composite resin restoration of MOD preparations. (G–H) Final prepared specimens for SSC and PFM crown placement, respectively Cavity surfaces were etched with phosphoric acid gel (Doradent, Iran) for 15 s, rinsed, and dried. James-2 bonding agent (Sarmeco, Switzerland) was applied for 20 s and light-cured for 30 s using a calibrated curing unit (Guilin Woodpecker Co., China) at 1000–1200 mW/cm². Cavities were restored incrementally with posterior A2 composite resin (Gradia Direct Posterior, GC Corporation, Japan). Restored specimens were stored in distilled water at 37 °C for 24 h prior to crown preparation. Tooth preparation For SSC group, teeth were prepared by reducing the occlusal surface for 1.5–2.0 mm. Proximal surfaces are reduced using a tapered fissure bur to create a smooth taper from occlusal to gingival, with mesial and distal contact points cleared for probe passage. Buccal and lingual reductions were reduced minimally for 0.5–1 mm (Fig. 1- G). For the PFM groups, the samples were prepared with high-speed diamond burs and 1.5 mm reduction was made from the occlusal surface, and 1 mm reduction was made from all other surfaces. The prepared teeth had 1 mm 360° chamfer margin, 6° convergence angle, 4 mm height, 8 mm mesial-distal width, and 6 mm buccal-lingual width (Fig. 1-H). Following tooth preparation, impressions were made using a two-step putty-wash technique. A heavy-body putty silicone (Coltene, Switzerland) was first seated over the prepared sample. After polymerization, relief space was created over the prepared tooth, and a light-body wash silicone (Coltene, Switzerland) was syringed onto the preparation margins before reseating the putty tray (Fig. 2). Impressions with voids or incomplete margins were discarded and repeated. Working dies were poured in type IV dental stone (Hersit 4, Heraeus Kulzer, Germany) for subsequent PFM crown fabrication. Two-step putty-wash impression technique following tooth preparation for PFM crown fabrication Crown fabrication and selection PFM crowns were fabricated using the conventional lost-wax casting technique. Two layers of die spacer were applied to the prepared dies prior to wax pattern fabrication to standardize a 20 μm cement space, consistent with conventional laboratory protocol. Wax patterns for the metal substructure were created at a uniform thickness of 0.5 mm. Patterns were invested and cast using a Ni-Cr alloy (61.4% Ni, 25.9% Cr, 11% Mo, 1.5% Si, 0.02% Mg, 0.02% Co). Cast frameworks were sandblasted with 50 μm Al₂O₃ particles. Vita VMK Master porcelain (Vita Zahnfabrik, Bad Säckingen, Germany) was applied following manufacturer’s instructions: a Vita OP3 opaque layer was fired directly onto the metal surface, followed by layered application, firing, and glazing of veneering porcelain using Vita Akzent Plus (Vita Zahnfabrik, Germany). Final restorations comprised a 0.5 mm metal substructure and 1.0 mm porcelain veneer, with standardized occlusal morphology fabricated as occlusal coping. For SSC groups, the appropriate crown size was selected by trial seating. Crowns were crimped to optimize marginal adaptation. Where a satisfactory fit could not be achieved with standard mandibular first molar crowns (Shinhung Co., Ltd., South Korea), crowns from adjacent sizes were trialed. In cases where standard sizing remained inadequate, larger-sized crowns from primary molar series were adapted. Internal fit evaluation Following crown fabrication and selection, internal fit was assessed prior to cementation using the silicone replica technique, as described by Tekin et al. [10]. A light-body polyvinyl siloxane material (Coltene, Switzerland) was injected into the internal surface of each crown, which was then seated onto the prepared tooth under finger pressure for five minutes until complete polymerization. The crown was subsequently removed, and the light-body silicone replica was stabilized by embedding in heavy-body silicone (Coltene, Switzerland) to facilitate sectioning and prevent distortion. Each replica was sectioned using a #11 surgical blade in the buccolingual and mesiodistal direction. Any distorted or ruptured replicas were discarded and repeated. Sections were aligned parallel to the optical axis of a digital microscope (Dino-Lite Pro, Anmo Electronics Corp, Taiwan) and examined at 10× magnification. At each cross-section, three measurement points were recorded along the cervico-occlusal axis of the light-body silicone layer (Fig. 3): DL0 and DL1, located at the occlusal and middle thirds of the axial wall respectively, representing the axial gap; and DL2, located at the cervical margin at the closest point between the crown margin and the finish line of the preparation, representing the marginal gap. Measurements were obtained from four surfaces (mesial, distal, buccal, and lingual) per crown. Each measurement was repeated three times and the mean value was recorded. The axial gap was calculated as the mean of DL0 and DL1, the marginal gap was defined as DL2, and the total internal gap was calculated as the mean of all three points. All measurements were performed by a single experienced operator. Assessment of internal fit using the silicone replica technique. Following sectioning of the replica, three measurement points were recorded along the cervico-occlusal axis of the light-body silicone layer at each cross-section under 10× magnification (Dino-Lite Pro, Anmo Electronics Corp, Taiwan): DL0 at the occlusal third of the axial wall, DL1 at the middle third of the axial wall, and DL2 at the cervical margin, representing the marginal gap Fracture resistance evaluation Following internal fit assessment, all crowns were cemented with glass ionomer cement (Vincismile, China) according to the manufacturer’s instructions and stored for 24 h prior to fracture testing. A vertical compressive load was applied to the center of the occlusal surface using a universal testing machine at a crosshead speed of 0.5 mm/min until fracture occurred. The maximum force at fracture was recorded in Newtons. Fracture mode was classified as repairable (crown distortion or partial fracture permitting repair or replacement without irreversible tooth damage) or non-repairable (catastrophic failure involving crown, core, or tooth structure precluding further restoration). Statistical analysis Data were analyzed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Normality was assessed using the Shapiro–Wilk test. A two-way MANOVA was conducted to evaluate the simultaneous effects of crown type (SSC vs. PFM) and tooth condition (intact vs. restored) on total internal gap, axial gap, and marginal gap, followed by univariate between-subjects analyses for each dependent variable. For fracture resistance, two-way ANOVA was used to assess main effects and interactions. Post-hoc pairwise comparisons were performed using independent-samples t-tests with Bonferroni correction where significant effects were detected. The Chi-square test was applied to compare fracture mode distribution between groups. A significance level of α = 0.05 was adopted throughout. Post-hoc observed power was calculated within SPSS. The main effect of crown type yielded an observed multivariate power of 1.000 (Wilks’ λ = 0.265, F(3, 34) = 31.513, p < 0.001, partial η² = 0.735), confirming that the sample size of 10 specimens per group was adequate to detect the observed differences across all internal fit outcomes. The effect of tooth condition and the interaction term yielded observed powers of 0.172 and 0.378, respectively. Results Descriptive statistics of the internal gap, marginal gap and axial gap among different groups are presented in Table 1. A two-way MANOVA revealed a statistically significant main effect of crown type across all three fit outcomes (p < 0.001, partial η² = 0.735). The effect of tooth condition (intact vs. restored) and the crown type × tooth condition interaction was not statistically significant (p > 0.05). Between-subjects analysis confirmed the crown type effect for total internal gap (p < 0.001, partial η² = 0.500), axial gap (p < 0.001, partial η² = 0.526), and marginal gap (p < 0.001, partial η² = 0.584). SSCs demonstrated greater total internal, axial and marginal gap values compared with PFMs. Table 1Internal fit values (µm) and between-subjects effects of crown type and tooth conditionOutcomeSSC (Mean ± SD)PFM (Mean ± SD)F (1, 36) p η²IntactRestoredIntactRestoredTotal internal gap788.09 ± 195.66766.49 ± 243.74460.27 ± 118.99558.92 ± 171.4135.947< 0.001*0.500Axial gap1025.00 ± 271.58996.70 ± 172.14599.00 ± 80.95726.70 ± 107.4339.875< 0.001*0.526Marginal gap313.20 ± 44.46304.90 ± 62.03183.32 ± 41.29222.10 ± 38.2550.453< 0.001*0.584*Significant at α = 0.05Effect of tooth condition: p > 0.05 for all outcomesInteraction (crown type × condition): p > 0.05 for all outcomesF values reflect the main effect of crown type (two-way MANOVA; p < 0.001, partial η² = 0.735) Internal fit values (µm) and between-subjects effects of crown type and tooth condition *Significant at α = 0.05 Effect of tooth condition: p > 0.05 for all outcomes Interaction (crown type × condition): p > 0.05 for all outcomes F values reflect the main effect of crown type (two-way MANOVA; p < 0.001, partial η² = 0.735) Fracture resistance data are summarized in Table 2. PFM crowns demonstrated significantly greater total fracture resistance than SSCs under both intact and restored conditions (p < 0.001). When SSCs were compared with the PFM porcelain fracture load specifically, no significant difference was observed in intact teeth; however, in restored teeth, SSCs demonstrated significantly higher fracture resistance than the PFM porcelain layer (p = 0.001). Tooth condition had a significant effect on SSC fracture resistance, with restored SSCs withstanding greater loads than intact SSCs (p = 0.029). In contrast, neither the porcelain fracture load nor the total fracture resistance of PFM crowns differed significantly between intact and restored conditions. Table 2Comparing the average fracture resistance (N) of the control and experimental SSC and PFM crowns in different surfacesSubgroupSSC (Mean ± SD)PFM porcelain fracture (Mean ± SD)p value(SSC vs. PFM porcelain fracture)PFM total fracture (Mean ± SD)p value(SSC vs. PFM total fracture)Intact teeth3447.80 ± 1126.092489.40 ± 1167.700.0786814.70 ± 422.23< 0.001*Restored teeth4831.80 ± 1456.572314.0 ± 1065.250.001*6924.80 ± 666.09< 0.001*p value (intact vs. restored teeth)0.029*0.4810.440*Values lower than 0.05 represent a significant difference between the groups according to independent samples t-test Comparing the average fracture resistance (N) of the control and experimental SSC and PFM crowns in different surfaces *Values lower than 0.05 represent a significant difference between the groups according to independent samples t-test The distribution of repairable and non-repairable fractures is presented in Table 3. A statistically significant difference in fracture mode was observed between SSC and PFM groups (p < 0.001). Among SSCs, 80% of intact specimens and 40% of restored specimens sustained repairable fractures. In contrast, all PFM crowns fractured in a non-repairable manner. Table 3The prevalence of repairable and unrepairable fractures among the groupsFailureSSCPFMp valueRestoredIntactIntactRestoredUnrepairable621010< 0.001*Repairable4800Total10101010 The prevalence of repairable and unrepairable fractures among the groups Discussion The present study compared the internal adaptation and fracture resistance of SSCs and PFM crowns in both intact and extensively restored permanent first molars. PFM crowns demonstrated superior marginal and axial adaptation across all surfaces and conditions, while SSCs exhibited more favorable and repairable fracture patterns. Regarding internal adaptation, both crown types exceeded the 100–120 μm marginal gap threshold widely regarded as clinically acceptable [11]. The inferior adaptation of SSCs is an expected consequence of their prefabricated, standardized geometry, which cannot conform precisely to individual tooth anatomy and relies on crimping and cement thickness for marginal sealing [12]. Muñoz-Sánchez et al. [13] observed marginal adaptation defects in approximately 37.5% of 360 radiographically evaluated SSCs. Al-Haj Ali [14], on the other hand, reported a comparable marginal gaps across conventional SSCs, preveneered SSCs and zirconia crowns when they were cemented with glass ionomer cement. Importantly, achieving a marginal fit of 120 μm with SSCs is feasible only with custom-fabricated restorations and is not attainable with prefabricated crowns [3]. However, evidence suggests this risk of marginal discrepancy is real but manageable. SSCs showed the largest total median vertical marginal gap among prefabricated crowns, yet longitudinal data indicate acceptable survival ranging from 64% to 100% over 18 months, when appropriate size selection, crimping, and cementation protocols are followed [3, 15]. Chaipattanawan et al. [7] noted that only 40% of SSCs on young permanent molars survived five years without intervention. Marginal defects exacerbated by continued eruption or inadequate crimping were the primary cause of failure. PFM crowns, fabricated on individualized dies, naturally achieved closer adaptation. However, their marginal gap values in the present study, particularly in restored teeth, modestly exceeded the 120 μm threshold, a finding consistent with the variability inherent in conventional laboratory fabrication. Both conventionally fabricated PFM and CAD/CAM crowns have been shown to achieve Although both conventionally fabricated PFM and CAD/CAM crowns can achieve clinically acceptable marginal fit [16], the presence of an underlying composite restoration in the present study was associated with increased marginal discrepancy in PFM crowns, while SSC adaptation remained unaffected. This differential response suggests that the precision demands of laboratory crown fabrication render PFMs more sensitive to changes in prepared tooth geometry introduced by restorative buildups. The fracture behavior of the two crown systems differed substantially in both magnitude and clinical implications. Restored SSCs demonstrated significantly higher fracture resistance than intact SSCs, suggesting that composite core buildups enhance load distribution within the crown-tooth complex. This is consistent with evidence that composite core-supported SSCs enhance the fracture resistance of severely damaged teeth, with fracture loads in core-supported SSC groups reaching up to 5500 N [17], closely approximating the values observed in the present study. PFM crowns porcelain fracture was approximately 2300–2500 N, with no significant difference between intact and restored teeth. Similar to our findings, evidence suggests that porcelain fractures typically occur at 2000–3000 N, after which the metal coping may remain intact but compromised esthetically and functionally [18]. However, PFM total fracture resistance was substantially higher than SSCs in both conditions (> 6800 N), reflecting the reinforcing contribution of the metal substructure. However, this mechanical superiority was accompanied by exclusively non-repairable failures, whereas 40–80% of SSC failures were repairable. Normal masticatory forces in the molar region range from approximately 400 to 800 N, meaning that both crown types far exceeded functional loading thresholds; the clinically meaningful distinction therefore lies in failure mode rather than fracture load. This contrast in failure behavior has direct implications for patient management. The ductile deformation characteristic of SSCs allows energy absorption under load, protecting underlying tooth structure and permitting repair or replacement without irreversible damage. Similarly, El Shahawy and Azab [18] reported that prefabricated crowns on molar analogs exhibited comparable fracture loads to custom CAD/CAM crowns but more favorable failure modes and none of which caused underlying substrate fractures. Similarly, a 12-month clinical study comparing SSCs and zirconia crowns in children found higher fracture rates among zirconia restorations (10%) than SSCs (2%) [19]. Another study reported that fracture forces required for preveneered SSCs were greater than those for prefabricated zirconia crowns [20]. Therefore, for young patients with ongoing jaw development, parafunctional habits, or high occlusal loads, the repairable failure mode of SSCs represents a clinically meaningful advantage. Therefore, the findings support a staged restorative strategy, with SSCs as interim full-coverage restorations during the mixed dentition period followed by replacement with a definitive laboratory-fabricated restoration once skeletal maturity and occlusal stability are achieved. However, careful crimping and frequent follow-up visits are warranted. This study has certain limitations. As an in vitro investigation, it could not replicate the complex oral environment, including temperature fluctuations, cyclic masticatory loading, and saliva-mediated cement degradation, all of which influence long-term crown performance. The absence of thermomechanical aging may have resulted in an overestimation of both marginal integrity and fracture resistance. Another limitation was that this study evaluated the mechanical behavior of the entire crown-tooth complex rather than the crown material in isolation. While clinically relevant, this approach prevents attributing fracture resistance solely to the material itself. To decouple material properties from tooth variables, future research should utilize standardized resin dies or metal analogs. Furthermore, only one brand of SSC and one PFM fabrication protocol were evaluated; variations in alloy composition, ceramic layering technique, and cementation protocol may yield different outcomes. Future studies should incorporate aging protocols, larger and more diverse sample sizes, and alternative fabrication technologies, including CAD/CAM zirconia and 3D-printed restorations, to better approximate clinical conditions and validate the present findings. Furthermore, the post-hoc power analysis revealed that the study was not adequately powered to detect small effects of tooth condition on internal fit outcomes (observed power = 0.172). Long-term clinical trials comparing SSCs and PFM crowns on young permanent molars are warranted. Conclusions Within the limitations of this in vitro study, PFM crowns demonstrated superior marginal adaptation and overall fracture resistance compared with SSCs. However, SSCs exhibited more favorable fracture patterns and showed greater fracture resistance than the porcelain layer of PFM crowns in isolation. The adaptation of PFM crowns appeared more sensitive to prior composite restoration than SSCs. These findings suggest that SSCs may represent a practical interim restorative option for young permanent molars in mixed dentition, particularly in clinical scenarios where fracture repairability and tolerance of underlying restorative buildups are priority considerations. Definitive conclusions regarding long-term clinical performance await validation through prospective clinical trials with extended follow-up periods. Acknowledgements This article was derived from two master’s thesis (code: 14011532 and 14011419). The authors express their appreciation to the Vice-Chancellor of Research of Mashhad University of Medical Sciences for generously funding this effort. Authors’ contributions H.M. and F.K. conceptualized the main idea, designed the study, supervised the research process, and critically reviewed the manuscript. F.S., A.A., and H.BK. were responsible for data acquisition and data analysis. F.S. drafted the original manuscript. All authors read and approved the final version of the manuscript. Funding No funding was obtained for the present study. Data availability The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate All procedures were executed in compliance with the Helsinki Declaration. The current in vitro study protocol was authorized by the Ethics Committee of Mashhad University of Medical Sciences (code: IR.MUMS.DENTISTRY.REC.1401.117 and IR.MUMS.DENTISTRY.REC.1401.116). Informed consent form was obtained from the patients to use their extracted teeth for the research project. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.