Marginal gap evaluation of zirconia and metal short crowns with varying total occlusal convergence and auxiliary preparation features: an in-vitro study

Abstract

Introduction

Marginal adaptation is one of the important factors in determining the long-term success of dental restorations. Despite extensive research, conflicting evidence exists regarding the influence of material type, total occlusal convergence (TOC), and auxiliary preparation features on marginal gaps of definitive crowns.

Objectives

This in-vitro study aimed to evaluate and compare the marginal gaps of zirconia and metal short crowns fabricated using digitally across preparations with varying TOC angles (10° and 20°) and auxiliary features (groove and box).

Methods

Eighty short crown specimens were distributed across eight groups (n = 10 each): zirconia and metal crowns at 10° and 20° TOC, with auxiliary groove or box preparations. Mandibular first molar typodont teeth were prepared under standardized conditions. Zirconia crowns were milled via CAD-CAM (Ceramill Zolid PS), and metal crowns were fabricated from digitally milled wax patterns using the lost-wax technique (Wiron Light). Marginal gaps were measured at seven equidistant buccal and lingual, and five mesial and distal points under 100× stereomicroscopic magnification. Data were analyzed using one-way ANOVA, Tukey’s post-hoc test, unpaired Student’s t-test, and paired t-test (α = 0.05).

Results

One-way ANOVA showed a statistically significant difference in marginal gaps among all groups across all surfaces (p < 0.05). Metal crowns demonstrated significantly higher marginal gaps (145.71 ± 54.50 μm) than zirconia crowns (104.02 ± 50.19 μm) (p = 0.001). The zirconia crown with 20° taper and buccal groove (Z20G) showed the lowest marginal gap (71.6 ± 48.2 μm), while the metal crown with 20° taper (M20) recorded the highest (178.0 ± 61.5 μm).

Conclusions

Marginal gap of short crowns is significantly influenced by material type, TOC, and auxiliary preparation design. Zirconia crowns with 20° taper and buccal grooves demonstrated accurate marginal fit, with taper angle, and auxiliary features significantly influencing marginal gap accuracy.

Introduction

The long-term success of indirect restoration depends on its marginal and internal adaptation. Poor marginal adaptation can cause detrimental effects on the periodontium and tooth [1]. These can include plaque accumulation, microleakage, secondary caries, and alteration to oral microflora [2]. Moreover, poor internal retention can cause fracture of the restoration [2]. Even though the acceptable marginal fit is widely discussed in literature, there is no consensus on an acceptable marginal gap or discrepancy. Clinically acceptable threshold as mentioned in various studies ranges between 40 μm and 120 μm [3–5], variations can be reported depending on the measurement location, restoration types, and finish lines, with values between 8 μm and 200 μm [1, 6, 7].

The accuracy of marginal fit made from digital workflow and auxiliary features has been investigated by numerous studies, but the results are inconsistent [8–10]. While some studies reported significant improvement in the marginal fit of a crown manufactured by digital workflow, some supported the conventional method [11–13]. Some authors reported no significant difference in the marginal fit of a crown manufactured by digital and conventional workflows [14, 15].

Marginal fit is influenced by tooth preparation, convergence angle, auxiliary features, surface topography, and fabrication technique [7, 16, 17]. Goodacre et al. recommended a 10°–20° convergence angle and minimum axial heights of 4 mm for molars and 3 mm for anterior teeth [18], although others reported larger gaps under these parameters and favored auxiliary features such as grooves and boxes, particularly in compromised teeth [13, 19]. Mechanical preparation factors account for over 50% of restoration failures compared with 20% attributed to biological factors [20], yet their relative contribution alongside manufacturing variables remains debated.

Grooves and boxes are advocated where heavy occlusal forces are anticipated in molars [13, 19, 21] or clinical crowns are short [22, 23], as they increase cementation surface area; proximal grooves in particular improve retention, cement space, and marginal adaptation compared with boxes [22, 23]. Xu et al., however, found shoulder grooves and sharp line angles significantly worsened marginal gaps in short crowns [24]. Adaptation also varies with material and workflow: metal crowns remain widely used for their documented performance [16], while 3D printing and CAD-CAM address limitations of conventional techniques [17]. Comparisons between the two remain conflicting [25, 26], favoring conventional methods in some reports [27] and digital methods in others [19].

In recent years, patients’ demand for all-ceramic crowns has increased because of aesthetic concerns and awareness of biocompatibility. Zirconia, due to its high strength and biocompatibility, is widely used in dentistry and is fabricated with the CAD-CAM technique. Regarding the marginal gaps of zirconia restoration, conflicting results have been published, with some studies reporting acceptable marginal gaps while others report significantly more marginal gaps [11, 12]. Despite extensive literature, a systematic review reported low evidence for marginal gaps of full-coverage restorations [5]. The inconsistencies of current laboratory-based studies have reported inconclusive results [6, 16, 17]. Moreover, there is an inconsistency in the reports regarding the acceptable marginal gaps in the short crowns prepared by the digital and conventional methods using different materials. Therefore, the present in-vitro study aims to evaluate the accuracy of marginal gaps of two types of definitive crowns, metal and zirconia, at different Total occlusal convergence (TOC) with proximal box and buccal grooves. The null hypothesis generated was that there was no effect of material types and TOC with proximal box and buccal grooves on marginal gaps of short crown.

Material and method

The study protocol was developed and submitted for ethical approval to the scientific committee of the College of Dentistry, Jazan University (reference number – CODJU-2221I). The current in-vitro study was developed to evaluate the effect of mesial box and buccal groove at 10- and 20-degrees TOC on the marginal fit of short crowns.

The sample size was calculated using the G Power sample size calculator (Heinrich Heine University Düsseldorf, Düsseldorf, Germany) version 3.1.9.7. With the power analysis of 80%, measuring alpha at 5% and a medium effect size of 0.45 [28]. The calculated sample size was 80. The samples were then divided into 8 groups, and the process is presented in Fig. 1.

Fig. 1.

Groups and sample distribution

Fabrication of the samples

The typodont teeth (Practicon, Greenville, NC, USA), mandibular left first molar (36), was prepared to simulate short crown preparations following the established guidelines [13]. The occluso-cervical height of 3 mm and a deep chamfer margin of 1 mm were prepared on the sample tooth. Functional cusp reductions were finished in two planes. The first model was assigned a total occlusal convergence (TOC) of 10° (Fig. 2A), while the second had a TOC of 20° (Fig. 2B). All dental procedures were performed using a high-speed air rotor handpiece (KaVo do Brazil Ind. Com. Ltd., Joinville, SC, Brazil), securely mounted on a dental cast surveyor (3005 Surveyor Type A, Dental Farm). Preparation was accomplished utilizing a round-end fissure diamond bur (102 R Shofu) with tapers of 10° and 20°, as appropriate. Once the preparation was finished, the samples were randomly divided into groups, and a prepared tooth mesial box (n = 20) (Fig. 2C) and buccal groove (n = 20) of 0.5 mm was prepared (Fig. 2D).

Fig. 2.

Standardized preparation, design and fabrication of crown specimens. A Prepared abutment with 10° total occlusal convergence (TOC) and 3 mm axial wall height. B Prepared abutment with 20° total occlusal convergence (TOC) and 3 mm axial wall height. C Fabricated veneer crown with buccal groove and a mesial box. D Fabricated metal crown specimens with a buccal groove and a mesial box

Standardization of preparations was confirmed using a putty index sectioned in multiple directions (Affinis, Coltène/Whaledent). Anatomical contours were rounded off, avoiding sharp angles or lines. Before manufacturing the functional models, taper verification was conducted using a computer system comprising a stereomicroscope equipped with a USB CCD camera (Amscope, Irvine, CA, USA), a desktop computer, and compatible measurement software (Version 3.7.12924).

Modifications were made in master models to include grooves and boxes. 3-D printed replica models were generated with a consistent 20-degree taper to ensure uniformity in the preparation among the groups. Verification protocols match those used for typhodont teeth preparation. Model scanning was done using a Bench Top Scanner (model number 4, 3Shape, Copenhagen, Denmark). Working models were fabricated with various 3D printed resins (Denta Model, Asiga, Alexandria, Australia) via a digital light processing 3D printer (Asiga 3D printer, Alexandria, Australia, serial number 70B3D5362C6A, model number PN01233), set at 50-micrometer layer thickness, 2.975-second exposure time, and 0° printing orientation after scanning.

On one printed working model, a mid-buccal groove measuring 1 mm in width and depth, measuring 0.5 mm above the finish line, was created using a vertically oriented 1-mm diameter cylindrical diamond bur. Another printed model featured a proximal box on the mesial aspect, mirroring the groove’s details but with a width enlarged to 3 mm. All preparation and modifications were carried out by a single operator (A.P.). Working models were derived from these four master models with varying features.

A total of 40 zirconia and 40 metal working models were 3D printed (Asiga 3D printer, Alexandria, Australia, Serial Number: 70B3D5362C6A, Model Number: PN01233), each with varying tapers and auxiliary features. Resin dies were inspected for quality, scanned to create STL files, and processed. Margins of the definitive crowns were set with 50 μm cement space. Forty monolithic zirconia crowns (Ceramill Motion, Amann Girrbach GmbH, Germany) were manufactured in groups of ten using 3D-milling. The monolithic restorations were fabricated from Ceramill Zolid HT + PS (Amann Girrbach AG, Koblach, Austria), a pre-shaded 4 mol% yttria-stabilized tetragonal zirconia polycrystal (4Y-TZP; nominal flexural strength 1000 ± 150 MPa). Blanks were dry-milled (Ceramill Motion 2, Amann Girrbach AG) and sintered according to the manufacturer’s protocol (peak temperature ~ 1450 °C, holding time 120 min, heating rate 8 °C/min), with software compensation for sintering shrinkage applied before milling.

Forty nickel-chromium metal crowns (Wiron Light) were manufactured in group of ten following similar wax pattern fabrication for zirconia crown and lost wax technique for casting. The fit of each crown was checked on its model; any inaccurate crowns were replaced. After preparation, the prepared dies were allocated to the eight groups using a computer-generated random number sequence, with allocation performed by an investigator not involved in specimen measurement. Prior to measurement, each specimen was assigned a numeric code by an independent investigator, and the examiner recording marginal gaps was blinded to material and preparation-design allocation.

Measurement of marginal fit

Before cementation, the crowns (zirconia and metal) were placed on prepared models to measure the fit. The putty index was fabricated on the horizontal platform of a stereomicroscope to allow standardized measurements of the samples. Each specimen was placed under a stereomicroscope at 100× magnification. The prototype dies were made on axial and proximal surfaces with flat cuts to place a metallic ruler to obtain the measurement. Seven measurement points were recorded at the buccal and lingual sides (Fig. 3A) and five points at mesial and distal to measure marginal gaps (Fig. 3B). To achieve optimal visualization, an LED double gooseneck illuminator (VMLIHL-20, Vision Scientific, Westland, MI, USA) was utilized. Measurement software (Amscope Version No. 3.7.12924, Irvine, CA, USA) was used to calculate the linear gap between the crown and the finish line at different points. Finally, all the acquired data were entered into Microsoft Excel (version 21) according to the sample distribution amongst the groups. A single measurement was recorded at each of the 24 predetermined points per specimen by one calibrated examiner (A.P). To document reproducibility, 20% of specimens were re-measured after a two-week interval; intra-examiner reliability was excellent (ICC = 0.87).

Fig. 3.

A Seven-point measurements at equidistant points on the buccal surface in a stereomicroscope and mounted camera. B five-point measurement at equidistance points on the distal surface in a stereomicroscope and mounted camera

Statistical analysis

The data were collected and imported into Microsoft Excel, and statistical analysis was performed using SPSS software (version 20; SPSS Inc., Chicago, IL, USA). Descriptive statistics were calculated using the mean, median, and standard deviation (in µm). The marginal gap among all groups was analyzed using one-way ANOVA followed by Tukey’s post hoc test. Assessment of normality was calculated using Shapiro-Wilk tests, which showed statistical variations amongst the groups. Hence, the non-parametric Mann-Whitney U test was performed to compare the two materials. The level of statistical significance was set at 0.05 for all tests.

Results

Table 1 demonstrates the means and standard deviations (SDs) of the mean marginal gaps for all the groups. The result of one-way ANOVA found a statistically significant difference among the groups (P < 0.05) on all the surfaces. Overall, the marginal gaps were maximum (213.6 μm) for the mesial side in group 4 (M20 group) and minimum (57.3 μm) on the buccal side in Z20G group. Within the groups, the lowest gaps were found for the 20° taper with buccal groove preparation in zirconia crowns, whereas the highest values within groups were noted for the 20° taper metal crown group.

Table 1.

Comparison of mean marginal gap values (µm) among eight experimental crown groups at buccal, mesial, lingual, distal, and total surfaces by one-way ANOVA

Group	Buccal		Mesial		Lingual		Distal		Total
	Mean	± SD	Mean	± SD	Mean	± SD	Mean	± SD	Mean	± SD
Z10	82.1	20.7	129.4	52	100.1	56.9	96.4	25.3	102.1	29.8
M10	152.6	77.4	176	64.3	145.4	62.8	145.4	40	154.7	53.6
Z20	163.3	106.7	134.2	62.3	84.1	50.6	125.6	84.6	126.8	58.2
M20	145.3	56.9	213.6	116.3	184.3	53.1	168.8	68.3	178	61.5
Z20G	57.3	38.7	83	74.7	75.4	59.1	70.6	43.3	71.6	48.2
M20G	110.7	46.2	136.8	69.3	121.7	61	124	48	123.3	52.2
Z20B	97.6	53.3	117.6	51.3	103.9	54.5	143.6	67.5	115.8	49
M20B	119.4	36.4	116.3	31.4	125.1	50.2	145.8	48.5	126.6	35.8
Total	116.0	66.9	138.4	76.0	117.5	63.1	127.5	61.1	124.9	56.1
P value	0.002*		0.004*		0.001*		0.006*		0.001*

Z10 zirconia crown with 10⁰ taper, M10 Metal crown with 10⁰ taper, Z20 Zirconia crown with 20⁰ taper, M20 Metal crown with 20⁰ taper, Z20G Zirconia crown with 20⁰ taper and groove, M20G Metal crown with 20⁰ taper and groove, Z20B Zirconia crown with 20⁰ taper and box, M20B Metal crown with 20⁰ taper and box

One-Way ANOVA: *p < 0.05; Significant

Post hoc Tukey HSD analysis identified statistically significant differences in marginal gap values among the eight groups (p < 0.05) (Fig. 4). The M20 group demonstrated significantly higher marginal gap values than the Z10 group at the lingual surface (p = 0.027). Similarly, the M10 group demonstrated significantly higher marginal gap values compared with the Z20G group at the buccal surface (p = 0.015). The Z20 group differed significantly from the M20 group at the lingual surface (p = 0.004) and from the Z20G group at the buccal surface (p = 0.004). The most consistent and pronounced statistically significant differences were observed between the M20 and Z20G groups, which differed significantly across all measured surfaces (p ≤ 0.001).

Fig. 4.

Post-Hoc Comparison of the total mean marginal gaps between eight groups by Tukey’s HSD test *p < 0.05; Significant; Z10: zirconia crown with 10° taper; M10: metal crown with 10° taper; Z20: zirconia crown with 20° taper; M20: metal crown with 20° taper; Z20G: zirconia crown with 20° taper and groove; M20G: metal crown with 20° taper and groove. Z20B: zirconia crown with 20° taper and box; M20B: metal crown with 20° taper and box

The influence of taper angle and retention features on marginal gap values, combining both crown materials, is presented in Table 2. One-way ANOVA demonstrated statistically significant differences among the four configuration subgroups at the buccal (p = 0.011), mesial (p = 0.017), distal (p = 0.040), and total (p = 0.044) measurements. No statistically significant difference was identified at the lingual surface (p = 0.390). The 20° taper showed the highest mean marginal gap of 152.40 ± 63.93 μm, while the 20° taper with groove demonstrated the lowest mean total marginal gap of 97.44 ± 55.69 μm.

Table 2.

Comparison of marginal gap values (µm) by taper angle and proximal retention features across all groups by one-way ANOVA

Taper/Aux. features		Buccal	Mesial	Lingual	Distal	Total
10°	N	20	20	20	20	20
	Mean	117.3571	152.7000	122.7857	120.9000	128.4358
	Std. Deviation	65.93665	61.74063	62.78785	41.15938	50.16467
	Median	88.5714	143.0000	97.1429	119.0000	118.3575
	Minimum	51.43	68.00	55.71	44.00	79.29
	Maximum	292.86	300.00	287.14	208.00	242.22
20°	N	20	20	20	20	20
	Mean	154.2857	173.9000	134.2143	147.2000	152.4000
	Std. Deviation	83.74813	99.54100	72.05511	78.02941	63.92851
	Median	125.7143	156.0000	135.7143	131.0000	152.9288
	Minimum	75.71	42.00	27.14	30.00	49.72
	Maximum	398.57	504.00	264.29	346.00	305.43
20° with Groove	N	20	20	20	20	20
	Mean	84.0000	109.9000	98.5714	97.3000	97.4428
	Std. Deviation	49.73528	75.36431	63.06698	52.26360	55.69000
	Median	80.0000	102.0000	82.1429	109.0000	100.1788
	Minimum	7.14	20.00	20.00	14.00	20.43
	Maximum	210.00	272.00	214.29	186.00	217.36
20° with Box	N	20	20	20	20	20
	Mean	108.5000	116.9600	114.5000	144.7000	121.1653
	Std. Deviation	45.82101	41.40875	52.13235	57.22081	42.14104
	Median	113.5714	113.0000	99.2857	131.0000	113.2150
	Minimum	34.29	62.00	60.00	62.00	59.00
	Maximum	170.00	224.00	240.00	292.00	206.57
p-value		0.011*	0.017*	0.390	0.040*	0.044*

*p < 0.05; Significant

The comparison of marginal gap values between the two crown materials is presented in Table 3. Metal crowns demonstrated significantly higher marginal gap values than zirconia crowns across all measured surfaces. The overall mean total marginal gap for metal crowns was 145.71 ± 54.50 μm, compared with 104.02 ± 50.19 μm for zirconia crowns (p = 0.001). At the buccal surface, metal crowns recorded a mean marginal gap of 132.00 ± 56.92 μm versus 100.07 ± 72.85 μm for zirconia crowns (p = 0.002). The mesial surface demonstrated the largest absolute difference between the two materials, with metal crowns exhibiting a mean gap of 160.68 ± 82.71 μm compared with 116.05 ± 61.87 μm for zirconia crowns (p = 0.014). The most statistically significant differences were observed at the lingual and distal surfaces, where metal crowns recorded means of 144.14 ± 60.30 μm and 146.00 ± 52.70 μm, respectively, compared with 90.89 ± 54.47 μm (p < 0.001) and 109.05 ± 63.88 μm (p < 0.001) for zirconia crowns.

Table 3.

Comparison of mean marginal gap values (µm) between zirconia and metal crown groups by Mann–Whitney U test

Materials		Buccal	Mesial	Lingual	Distal	total
Zirconia	N	40	40	40	40	40
	Mean	100.0714	116.0500	90.8929	109.0500	104.0161
	Std. Deviation	72.84885	61.86584	54.47498	63.88289	50.19015
	Median	83.5714	114.0000	77.8571	103.0000	96.2863
	Minimum	7.14	20.00	20.00	14.00	20.43
	Maximum	398.57	252.00	257.14	292.00	233.22
Metal	N	40	40	40	40	40
	Mean	132.0000	160.6800	144.1429	146.0000	145.7058
	Std. Deviation	56.91971	82.71194	60.29997	52.69749	54.50463
	Median	117.8571	143.0000	135.7143	133.0000	137.5363
	Minimum	57.14	48.00	31.43	26.00	56.07
	Maximum	292.86	504.00	287.14	346.00	305.43
p-value		0.002**	0.014*	0.000**	0.000**	0.001**

*p < 0.05; Significant; **p < 0.001; Highly significant

Discussion

The present study evaluated the marginal gaps of definitive crowns with preparation tapers of 10° and 20°, as well as a 20° taper with auxiliary features (groove and box). The null hypothesis was rejected, as significant differences in marginal gaps were observed at several surfaces among the groups. Previous studies have reported two main approaches for measuring marginal gaps: (1) invasive or destructive methods, such as cementation followed by cross-sectioning and film thickness evaluation; and (2) non-invasive methods, including direct viewing, superimposition, replica techniques, and scanning methods [5, 6, 29]. In the present study, the marginal gaps of zirconia and metal crowns were evaluated before cementation; therefore, invasive methods were not employed. Instead, a non-invasive technique involving microscopic evaluation combined with image-analysis software was used.

Studies have reported that short clinical crowns pose a challenging task for clinicians in terms of achieving complete retention and complete crown coverage. During mastication, the vertical and horizontal forces are applied to the crown, which may result in dislodgment. Hence, studies have supported the incorporation of auxiliary features like proximal box and grooves. Weed et al., in their study, compared marginal gaps and retention following auxiliary features (grooves and boxes) and reported that box preparation provided clinically acceptable marginal gaps and retentive features [23]. However, Vinaya et al. reported that proximal grooves provided better retention and acceptable marginal gaps [22]. With the addition of auxiliary features in short crowns with increased TOC, both the preparations showed acceptable marginal gaps. Conversely, in the study by Alquatani et al., proximal boxes with short crowns prepared with 10-degree TOC showed acceptable marginal gaps compared to grooves and boxes with increased TOC [2]. In the current study, metal crowns prepared at 20-degree taper showed the highest marginal gaps. This discrepancy may be attributed to the fabrication technique.

The result of the current study measured significant difference between partial (digital + conventional) and complete digital workflow on the marginal fit. In a systematic review with low evidence, similar performance was reported on the marginal and internal fit of full coverage crowns prepared by both digital and conventional techniques [5]. The lost-wax casting technique used for metal crown fabrication introduces multiple potential sources of dimensional error, including wax pattern distortion, investment expansion variability, thermal shrinkage during solidification, and post-casting finishing inaccuracies [27]. Distortion in any of the fabrication steps can lead to increase in marginal discrepancy. The CAD-CAM milling process used to fabricate zirconia crowns operates from digitally captured STL files with sub-micron scanning precision, enabling highly reproducible and consistent crown preparations [30]. Additionally, zirconia undergoes predictable linear sintering shrinkage during the final densification stage, and modern CAD-CAM software digitally compensates for this known shrinkage factor prior to milling, ensuring that the final sintered crown closely fits to the prescribed dimensions [10]. Boitelle et al. reported that CAD-CAM fabricated restorations demonstrated significantly better internal and marginal fit compared to conventionally cast restorations due to the elimination of human-dependent fabrication variables [5]. These mechanistic differences provide a strong theoretical basis for the superior marginal adaptation of zirconia crowns observed across most groups in the present study.

A significant difference was observed in the marginal gaps across the eight groups (p = 0.001). The marginal gap was highest at the mesial side of metal group with 20-degree taper and minimum at the buccal side of zirconia group with 20-degree taper and buccal groove. Maximum published studies measured no significant differences between the marginal gaps across the group, except for 2 studies where the mesial gap was reported to be higher (due to higher firing shrinkage) [14, 25]. Holmes et al. emphasized that marginal gaps below 100 μm are associated with reduced microleakage and better clinical longevity, while gaps exceeding 200 μm significantly increase the risk of restoration failure [31].

The 20-degree taper with buccal groove reported the minimum marginal gap across all the subgroups. Similar results were obtained in the studies by Bowley et al. [32] and Vinaya et al. [22] where marginal gaps were reduced with incorporation of proximal grooves in molar crown preparation. Sayed et al. in their in-vitro evaluation reported significant improvement in marginal integrity of zirconia crown prepared at 20-degree with proximal groove. Harita et al. mentioned the incorporation of additional grooves at both sides of crown to reduce marginal gaps [33]. Conversely, in the study by Naik, no significant differences in marginal adaptation were observed with incorporation of additional auxiliary features [34]. Weed et al. compared various auxiliary features and supported the box preparation [23]. The findings of the present study therefore suggest that, for clinicians treating short crown, the combination of zirconia material with a 20° TOC and proximal groove preparation offers the most favorable marginal adaptation.

The mean marginal gaps across the groups in the current study were within a clinically acceptable range, except for metal group with 20-degree taper. However, the values of marginal gaps have not been standardized yet. Studies have reported that the clinically acceptable marginal gap should be between 40 and 170 μm [21, 22, 35]. An in-vitro study stated that there was no conclusive evidence on the optimal fit of the crowns, and the values ranged between 7.5 μm and 206.5 μm [14]. The difference in comparing the results of various research arose due to diversity in methodological approaches, material selections, crown preparation, and the difference in the terminologies to define and quantify marginal gaps. Moreover, no general opinion on the number of measurements has been found in the literature.

In the current study, the authors have measured seven points on the lingual and buccal side and five points on the mesial and distal side. The mean for each group was reported as 24 measurements per specimen and 1920 in total for all 80 crowns. The mean for each group was derived from 240 measurements for marginal discrepancy. Similar points were measured in the studies by Murgi et al. [19] and Heboyan et al., [25] whereas Gassino et al. [15] used 18 points at the buccal side and Nawafleh et al. [20] used 50 points at mesial and lingual sites, affecting the findings. Moreover, several in vitro studies aimed to provide reproducible and standardized test parameters that can be followed by other researchers; however, in vitro studies failed to provide clinical simulations, limiting the reliability of their results [26, 27, 30]. Moreover, the largest gaps (mesial 504 μm; total 305 μm) occurred within the M20 group and reflect the wider dimensional variability due to incorporation of lost-wax casting at 20° taper. Statistical analysis did not identify them as outliers, hence, they were therefore retained, and medians are reported alongside means to convey their influence on central tendency and dispersion.

Metal crowns demonstrated significantly higher marginal gap values than zirconia crowns, with the most significant differences at the lingual and distal surfaces. Huang et al., in their in-vivo study, reported better marginal fit of metal ceramic crowns manufactured using the laser sintering method than a CAD-CAM fabricated zirconia crown [36]. Conversely, Al Hamad et al., [4] and Al-Kheraif et al. [37] reported no significant differences in marginal fit of metal and zirconia crowns prepared at different TOC. Seelbach et al. reported no significant difference in marginal fit of three different types of crowns (LAVA, zirconia, and metal) using direct scanning and lost-wax techniques at different TOC [38]. However, studies have shown that metal and zirconia frameworks prepared by direct scanning techniques show significant marginal discrepancies [24, 39]. Recently published in-vitro studies reported that CAD-CAM fabricated metal and zirconia restorations demonstrated better marginal fit and clinical outcomes over conventionally prepared ones [13]. Previously published literature does not coincide with the findings of the present study. This discrepancy can be explained by several factors associated with CAD-CAM milling process. At steeper taper angles, the milling bur must navigate more acute geometric transitions along the axial walls, which can result in slight over-milling or rounding of internal line angles, thereby reducing the precision of the marginal adaptation [8, 26, 39]. Furthermore, the wider divergence associated with a 20° taper increases the mechanical load per unit area on the milling tool during the finishing pass, potentially introducing micro-vibrations that affect the accuracy of the marginal seating surface [19]. Discrimination between these two taper angles may also be less accurate in bench-top scanning at wider divergence angles due to specular reflection artifacts during the 3D scanning of die surfaces [14].

All marginal gap measurements in the present study were conducted prior to cementation, a methodological decision that has important implications for the clinical applicability of the reported values. The cementation process introduces several factors that can significantly alter the final marginal gap of a seated crown, including hydraulic pressure generated during cement seating, the viscosity and film thickness of the luting agent, elastic deformation of the die under occlusal seating pressure, and the degree of cement space provided in the restoration design [23, 24]. Studies evaluating pre- versus post-cementation marginal gaps have consistently reported that cementation increases marginal gap values by a mean of 20–60 μm, depending on the cement type and application technique [22, 39]. Piwowarczyk et al. reported that the use of resin-modified glass ionomer cements produced smaller post-cementation gap increases compared to conventional zinc phosphate cements due to their lower seating pressure requirements and better flow characteristics [40]. The 50 μm cement spacing incorporated into the crown designs in the present study was intended to partially compensate for this phenomenon; however, the actual post-cementation gaps may still differ from those reported. Clinicians and researchers should therefore interpret the values in the present study as representing the intrinsic fabrication accuracy of each group, rather than the definitive clinical marginal gap that would be observed after luting.

Limitations

There are several limitations in this study. Firstly, the in-vitro design does not replicate the oral environment and provides insufficient information to be applied in clinical practice. While the current study followed strict guidelines to reduce bias, the inconsistency can be reported in measuring parameters, and data could be misread by the operator and researchers. Additionally, the study faced constraints like image resolution, inaccuracy in crown placement, and differences in points of measurement. The current study has not evaluated marginal gaps after cementation, which could have altered the results. Nonetheless, the findings of the current study indicate clinically acceptable marginal gaps in the crowns fabricated by both zirconia and metal substrates under different preparation procedures. A restoration may show acceptable marginal adaptation yet poor internal fit; the absence of internal-fit data is therefore a limitation, and future studies using replica or micro-CT methods are required.

Furthermore, long-term clinical studies with a follow-up period of at least five years are required to understand the durability, failure rates, and patients’ satisfaction associated with these crowns. Researchers should focus on the role of auxiliary features and preparation methods on the retention of the short crown, as the data on this is limited. The impact of cementation on the crown fit and strength evaluation is required.

Material type and fabrication technique were confounded by design, zirconia crowns were CAD-CAM milled, whereas metal crowns were produced by the lost-wax casting route. The observed differences therefore reflect the combined effect of material and manufacturing methods and cannot be attributed to material composition alone. This is a major limitation, and future work should isolate the two factors.

Conclusion

The present in-vitro study demonstrated that marginal adaptation of short crowns is significantly influenced by material type, total occlusal convergence, and auxiliary preparation design. CAD-CAM fabricated zirconia crowns exhibited superior marginal adaptation compared to cast metal crowns across most groups, with buccal groove preparation at 20° TOC providing a favorable result. CAD-CAM zirconia crowns showed lower marginal gap values than cast metal crowns. These values represent fabrication accuracy and require confirmation after cementation and in clinical service before design recommendations can be made. Both materials produced clinically acceptable marginal gaps when appropriate preparation designs were employed. Long-term clinical studies are required to validate these in-vitro findings.

Authors’ contributions

Amit Porwal: Study design/planning; Data collection/entry; Data interpretation; Preparation of manuscript. Mohammed E. Sayed: Study design/planning; Data collection/entry; Data analysis/statistics; Preparation of manuscript. Kulashekar Reddy Nandalur: Study design/planning; Data analysis/statistics; Preparation of manuscript. Naveen Reddy Ramireddy: Study design/planning; Data analysis/statistics; Preparation of manuscript; Shekhar Gupta: Data collection/entry; Data analysis/statistics; Hussain D. Alsayed: Data analysis/statistics; Preparation of manuscript; Ahid Amer Alshahrani: Study design/planning; Data collection/entry; Data interpretation; Saeed M. Alqahtani: Preparation of manuscript; Mazen Ahmed Qadi: Final revision of manuscript; figures and tables cross checking: Saad Saleh AlResayes: Data collection/entry; Data interpretation.

Funding

No external funding was received in this project.

Data availability

Data will be available at the request from corresponding author.

Declarations

Ethics approval and consent to participate

The study protocol was developed and submitted for ethical approval to the scientific committee of the College of Dentistry, Jazan University (reference number – CODJU-2221I).

Competing interests

The authors declare no competing interests.