3D Printed Posterior Connector Dimensions’ Effect on Fracture Properties of Provisional Two-Unit Fixed Dental Prostheses

Abstract

This in vitro study aims to investigate the fracture properties of 3D-printed resin provisional material designed with different connector dimensions for two-unit fixed dental prostheses (FDPs). The master model was digitally designed following Shillingburg’s all-ceramic restoration tooth preparation guidelines and milled from aluminum. Four two-unit FDPs with different connector dimensions were designed: 2 × 3 mm, 3 × 3 mm, 3 × 4 mm, and 4 × 4 mm (width × length) (Groups A, B, C, and D, respectively; n = 10 for each group). These specimens were printed using 3D-printed resin material (Detax FREEPRINT^® temp). Forty specimens were subjected to a three-point test using a universal testing machine until fracture. The failure mode was examined under a stereomicroscope. The Kruskal–Wallis test at α = 0.05 revealed non-significant differences in fracture resistance load but significantly different elastic modulus, yield strength, and compressive strength (p = 0.061, p < 0.001, p < 0.001, and p < 0.001, respectively) among the different groups. The 2 × 3 mm connectors had higher means of modulus, yield strength, and compressive strength compared to the other groups. The study found that the maximum load causing fractures in 3D-printed provisional material connectors was consistent, regardless of connector cross-section variations. The 2 × 3 mm group performed best, while the 4 × 4 mm group performed worst.

1. Introduction

In prosthodontics dental rehabilitation, fixed dental prostheses (FDPs) are the most common choice in dental clinics, enhancing patient desire and satisfaction [1]. FDPs are any prosthesis fixed to natural teeth or dental implants the patient cannot remove [2]. The main goals of fabricating FDPs are restoring patients’ function, aesthetics, and comfort [3]. Furthermore, FDPs contain multiple components. The main FDP components are pontic, retainer, connector, and abutment [3]. The FDP components are designed, constructed, and manufactured in the dental laboratory. These laboratory steps should be finished before FDPs are cemented in the oral cavity [3].

According to the material used, these FDPs could be a definitive or provisional prosthesis [3]. Several materials can be used to fabricate the FDPs, such as ceramic, porcelain fused to metal, lithium disilicate-based ceramics, zirconia, and densely sintered alumina [4]. Nowadays, the most common material is ceramic, which has extraordinary properties. It has high aesthetic characteristics and good mechanical properties. This is in addition to more biocompatibility than metal fused to the porcelain prosthesis [4].

Although dental implants have a higher success rate than FDPs, there are still some limitations to placing them for all patients [5]. Additionally, dental implant treatment has some contraindications, such as periodontal disease, poor oral hygiene, heavy smoking, parafunctional habits like bruxism, and certain systemic diseases that affect the periodontium, bleeding control, or healing after surgical procedures [6]. Therefore, FDPs remain a valid and reliable treatment option in prosthodontics.

FDP abutments should be protected between dental visits using provisional FDPs [3]. A dental provisional prosthesis is used to improve the current condition of the oral cavity. It is a dental interim prosthesis used for a limited period before a permanent dental prosthesis is replaced [2]. A dental provisional prosthesis is designed to enhance esthetics, provide stabilization, and facilitate function [2]. It aids in avoiding gingival growth or inflammation of the periodontium and eliminates the need to remake the definitive impression [3].

Provisional FDPs could be fabricated using different provisional materials [7,8]. Polymethyl methacrylate (PMMA) is one of these materials that is widely used in dental applications [8]. These materials could be manufactured conventionally as cold-cured or heat-cured PMMA materials. In addition, these provisional FDPs could be manufactured digitally using subtractive or additive manufacturing technology [7,8].

Subtractive technology is defined as the machining process to remove material from a material block [2]. 3D-printing technology, which is known as additive manufacturing technology, is a general term for a manufacturing approach that transfers an object from liquid to solid structure by curing layer over layer until it reaches the final 3D structure [9,10]. These objects were designed primarily with specific designing software. Then, this 3D design is manufactured from varying resin materials using a 3D printer machine [9,11].

In dentistry, 3D-printing technology has a multi-advantage covering rapid manipulation, high precision, lower cost, simplicity and flexibility of the complex workflow, reduced amount of wasted raw material, good marginal adaptation, and reduced internal gap of dental prosthesis compared to other manufacturingNb methods [11,12,13,14,15]. A 3D-printed resin is one of these materials that is widely used in prosthodontics [7,11,12]. Nowadays, 3D-printed resin material is used for provisional crowns or FDPs [7].

The connector is a part that unites the retainer(s) and pontic(s) of FDPs, which could be a rigid or nonrigid connector [2]. In addition, the connectors should be large enough to improve the strength without impeding plaque control [3]. However, it is considered the weaker portion of FDPs. So, the connector dimension and type of connector are important variables that play a critical role in the longevity of the FDPs [16,17,18,19]. Flexural strength, fracture toughness, color stability, yield strength, and hardness are the most measurable properties in the literature regarding dental application [20]. In terms of mechanical parameters of any material, fracture strength is defined as the ability of a material to resist failure and is designated according to the mode of load application, such as compressive or tensile [21].

Although all previous studies used an all-ceramic material to measure the most suitable connector dimension, the literature lacks information on adequate 3D Printed Posterior Connector Dimensions. Therefore, this in vitro study aims to investigate the fracture properties of 3D-printed resin provisional material designed with different connector dimensions for two-unit FDPs.

The null hypothesis is that the 3D-printed posterior connector dimensions do not influence the fracture properties of provisional two-unit FDPs.

2. Materials and Methods

2.1. Master Model Design

Two adjacent teeth were fabricated digitally to simulate a clinical situation of maxillary posterior teeth to be restored with two-unit FDPs. Teeth were prepared digitally using (Autodesk Fusion 360 Professional, Autodesk, San Francisco, CA, USA) software following Shillingburg tooth preparation guidelines for all-ceramic restoration, with rounded line angles, deep chamfer finish line, teeth dimensions simulate the average measurement of maxillary first molar for both abutments and occluso-gingival preparation height 5.5 mm [22,23].

2.2. Master Model Manufacturing

The prepared model was fabricated using Aluminum material with CAM software MasterCam 2023 (Mastel Aluminium-Halbzeuge GmbH, Talheim, Germany) to generate the tool path for a 3-axis CNC milling machine. Then, the metal model is polished with a metal polish compound. Finally, the metal model was scanned using a Ceramill Map 400+ scanner (AmannGirrbach, GmbH for Ceramill Map 400 Scanner, Koblach, Austria), and the master digital model was exported as a Standard Tessellation Language (STL) file.

2.3. FPDs Design

Two-unit FDPs were designed on the master model STL file using Exocad software (DentalCAD 3.1 Rijeka, Exocad, Darmstadt, Germany). The two-unit FDPs were connected in the design with a triangular connector shape. The connector had four different dimensions selected based on the manufacturer’s recommendations (4 × 4 mm) and previous literature for the other groups [16,18,24,25,26].

Table 1. The tested groups and their connector’s dimensions.

Group	Width (mm)	Height (mm)	Cross-Sectional Area (mm2)
A	2	3	8.16
B	3	3	10.30
C	3	4	13.33
D	4	4	15.08

shows the sizes of the tested connectors. Figure 1 shows the design of the group D connector.

2.4. FPDs Manufacturing

Ten specimens from each design (a total of 40 specimens) were printed using a 3D mixture of acrylic/methacrylic resin material (Detax FREEPRINT^® temp, Detax GmbH, Ettlingen, Germany) and a Straumann P30+ 3D printer (Straumann, Andover, MA, USA). After printing, the printed specimens were washed, cleaned, and post-cured according to the manufacturer’s recommendations. Finally, supporting structures were removed, and specimens were finished and polished. However, all specimens were designed and printed by a single examiner.

2.5. Specimens Testing

This study evaluated and characterized connector failure in two-unit FDPs. Mechanical performance was assessed by measuring fracture load, compressive strength, yield strength, and failure modes under in vitro static loading. All forty specimens were examined under the stereomicroscope for cracks or fractures. Any specimen with cracks or fractures was excluded, and another specimen was printed using the same protocol. All groups were cemented to the metal model using temporary cement, Temp-Bond™ Clear (Kerr Dental, Orange, CA, USA). Each specimen of these groups was subjected to a three-point test using a universal testing machine, Instron 5969 (Instron, MA, USA). The metal model is positioned in the center of the universal testing machine at a 0° angulation. All specimens were subjected to vertical load until visually fractured in a universal test machine. The load was transferred through a 2.5 mm diameter steel ball at a 0.5 mm/minute crosshead speed at one point in the connector area of two units of FDPs.

2.6. Specimens Evaluation

The mode of failure was examined by one experienced examiner and classified according to the mode of failure mentioned in

Table 2. The mode of failure that occurred in the specimens.

	Crown Fracture
Type	First (Mesial)	Second (Distal)	Fracture Direction	Connector Fracture
1	Yes	No	One direction (mesial to distal)	No
2	Yes	No	Multiple directions	No
3	Yes	No	One direction (mesial to distal)	Yes
4	Yes	No	Multiple directions	Yes
5	No	Yes	One direction (mesial to distal)	No
6	No	Yes	Multiple directions	No
7	No	Yes	One direction (mesial to distal)	Yes
8	No	Yes	Multiple directions	Yes
9	Yes	Yes	One direction (mesial to distal)	No
10	Yes	Yes	Multiple directions	No
11	Yes	Yes	One direction (mesial to distal)	Yes
12	Yes	Yes	Multiple directions	Yes

. Figure 2 shows an example of a type 1 fracture.

2.7. Statistical Analysis

Descriptive statistics were performed according to the assumptions for statistical tests: normality, independence, homogeneity of variance, and outliers; the Kruskal–Wallis test at α = 0.05 was used as a statistical test.

3. Results

Descriptive statistics were performed. Means, medians, and standard deviations are shown in

Table 3. Median, mean, 95% confidence interval, standard deviation (SD), and sample size for the maximum load (N), elastic modulus (MPa), yield strength (MPa), and compressive strengths (MPa) by connectors’ dimensions.

	Group	A	B	C	D
	n	10	10	10	10
Maximum Load (N)	Median	1251.32	1136.06	1016.78	853.59
	Mean	1176.9	1159.6	1161	904.7
	95% Confidence Interval for Mean	989.22–1364.64	1021.50–1297.71	920.50–1401.48	742.61–1066.85
	SD	262.4	193.06	336.19	226.6
Elastic Modulus (MPa)	Median	533.92	451.09	440.57	394.83
	Mean	537.62	451.9	421.79	359.5
	95% Confidence Interval for Mean	488.91–586.34	418.79–485.01	377.27–466.32	302.09–416.91
	SD	68.1	46.28	62.24	80.25
Yield Strength (Offset 0.2%) (MPa)	Median	104.44	82.21	67.48	48.00
	Mean	94.14	84.78	67.77	37.98
	95% Confidence Interval for Mean	67.98–120.29	79.72–89.85	61.89–73.65	20.81–55.15
	SD	36.56	7.08	8.22	24
Compressive Strength (MPa)	Median	153.35	110.30	76.28	56.61
	Mean	144.23	112.58	87.09	60
	95% Confidence Interval for Mean	121.23–167.24	99.17–125.99	69.05–105.14	49.25–70.75
	SD	32.16	18.74	25.22	15.03

. Factors were Independent. According to the Shapiro-Wilk test used for normality assessment, there were few significant values for the dependent variables and groups p < 0.05. Therefore, the data is not normally distributed and considered using a nonparametric variant of the ANOVA test, the Kruskal–Wallis’s test. The Kruskal–Wallis test was conducted to determine whether there were significant differences in fracture resistance load by connectors’ dimension.

The homogeneity of variance is a key factor in the study analysis. The data were homogenous for the maximum load and modulus (p > 0.05), which is favorable for the study analysis. However, for yield and compressive strength, the data were heterogeneous (p < 0.05), potentially affecting the conclusions. Preliminary screening for outliers showed few outside the favorable range recommended by Tukey (1977), Hoaglin (1983), and Hoaglin (1985) [27,28,29]. The outliers were managed by imputation, replacing the outlier in question with the median according to the method described by Tukey (1977), Hampel (1986), and Rousseeuw (1987) [29,30,31]. After managing outliers, there were no outliers in maximum load and compressive strengths, while few outliers were in the modulus and yield strength dependent variables within the acceptable range, as shown in Figure 3.

The independent-sample Kruskal–Wallis test was as follows: The maximum load (N) was 7.374, df = 3, p = 0.061; The modulus (MPa) was 22.365 df = 3, p < 0.001; for yield strength (Offset 0.2%) (MPa) was 24.221, df = 3, p < 0.001; for compressive strength (MPa) was 26.770, df = 3, p < 0.001.

Figure 4 and

Table 4. Pairwise comparisons of groups.

	Test Statistic	Sig.	Adj. Sig. a
Maximum Load (N) across Group
Group D-Group C	10.30	0.05	0.29
Group D-Group B	11.80	0.02 *	0.14
Group D-Group A	12.30	0.02 *	0.11
Group C-Group B	1.50	0.77	1.00
Group C-Group A	2.00	0.70	1.00
Group B-Group A	0.50	0.92	1.00
Elastic Modulus (MPa) across Group
Group D-Group C	8.00	0.13	0.76
Group D-Group B	13.10	0.01 *	0.07
Group D-Group A	24.10	<0.001 *	0.00
Group C-Group B	5.10	0.33	1.00
Group C-Group A	16.10	0.00 *	0.01
Group B-Group A	11.00	0.04 *	0.21
Yield Strength (Offset 0.2%) (MPa) across Group
Group D-Group C	11.00	0.04 *	0.21
Group D-Group B	20.25	<0.001 *	0.00
Group D-Group A	23.35	<0.001 *	0.00
Group C-Group B	9.25	0.08	0.46
Group C-Group A	12.35	0.02 *	0.11
Group B-Group A	3.10	0.55	1.00
Compressive Strength (MPa) across Group
Group D-Group C	10.10	0.05	0.32
Group D-Group B	18.80	<0.001 *	0.00
Group D-Group A	25.50	<0.001 *	0.00
Group C-Group B	8.70	0.10	0.58
Group C-Group A	15.40	0.00 *	0.02
Group B-Group A	6.70	0.20	1.00

Each row tests the null hypothesis that the group 1 and group 2 distributions are the same. Asymptotic significances (2-sided tests) are displayed. * The significance level is 0.050. ^a Significance values have been adjusted using the Bonferroni correction for multiple tests.

show the pairwise comparisons of group results.

Table 5. The distribution of fractures is in counts according to the location of the specimens.

			Maximum Load (N)		Elastic Modulus (MPa)		Yield Strength (Offset 0.2%) (MPa)		Compressive Strength (MPa)
Group	Fracture Types	Count	Mean	SD	Mean	SD	Mean	SD	Mean	SD
A	6	2	1378.65	56.36	543.74	181.68	49.31	59.98	168.95	6.91
	8	1	1509.82	.	536.94	.	122.54	.	185.03	.
	11	3	918.70	206.29	548.47	35.82	86.59	25.45	112.59	25.28
	12	4	1186.52	215.08	526.60	41.82	115.11	11.31	145.41	26.36
B	4	1	1241.98	.	448.84	.	93.25	.	120.58	.
	10	1	1606.83	.	462.14	.	82.46	.	156.00	.
	12	8	1093.41	115.64	451.00	52.32	84.02	7.26	106.16	11.23
C	9	1	961.81	.	368.82	.	71.59	.	72.15	.
	10	1	849.49	.	447.29	.	62.87	.	63.73	.
	11	2	1016.78	76.13	360.07	123.99	60.72	10.35	76.28	5.71
	12	6	1294.17	381.23	446.95	30.40	70.30	8.05	97.09	28.60
D	1	5	779.57	113.26	368.12	66.82	40.19	21.78	51.70	7.51
	11	2	783.20	147.54	308.17	128.34	23.76	28.93	51.94	9.79
	12	3	1194.35	121.81	379.34	91.69	43.77	30.99	79.20	8.08

shows the distribution of fractures according to location in the specimens. Fracture type 12 (multiple-direction fracture) was the most common type for all groups. Fracture types 2, 3, 5, and 7 had not occurred in any specimens.

4. Discussion

The null hypothesis was retained for the maximum load across the different tested groups. At the same time, it was rejected for elastic modulus, Yield Strength (Offset 0.2%), and Compressive Strength when the significance level was p < 0.001.

Temporary or interim fixed prosthodontics are commonly used as a step towards fabricating final fixed prostheses [3,22]. 3D-printed temporaries or milled PMMA can serve as an interim treatment option [32]. One observation from an author during the clinical experience was that most fractures in 3D-printed temporary prostheses occurred in the connector area. This prompted the authors’ interest in evaluating the factors that may lead to connector failures. Connectors are between two adjacent retainers and between a retainer and a pontic [2]. To characterize the effects of chewing and mastication on connectors, we begin with the simplest form—connectors between two crowns, namely, splinted crowns. Then, we will progress to more complex connector designs in future studies. Long-span bridges or fixed procedures rely on splinting the retainers at the end of the prosthesis to ensure a strong structure. The dimensions of the connectors are crucial factors that can affect the outcome of fixed prosthodontics treatment [22].

This study differs from previous studies in isolating the connector area and testing the effects of direct static loading on the connector area of two splinted crowns [16,18,19,24,25,26,33,34].

The methodology utilized two adjacent abutments on metal dies that were not previously used in any study. We employed a metal die made of milled aluminum, with dimensions standardized according to Shillingburg’s recommendations and average tooth dimensions [22,23]. The production of these 3D milled dies uses 3D software Fusion 360 (Autodesk Fusion 360 Professional, Autodesk, San Francisco, CA, USA), making it more accurate than other methods of fabricating metal dies to specified dimensions. The two connected retainers were also designed with the 3D software Exocad (DentalCAD 3.1 Rijeka, Exocad, Darmstadt, Germany), which was used to select the measurements of the connector area during the design process.

Testing using the Instron machine (Instron 5969, Norwood, MA, USA) utilized a 2.5-mm three-dimensional generated rod that simulates the loading of a cusp area directly on the connector. The loading was configured at a 0-degree angle from the connector’s vertical axis.

In the literature, the connector area is commonly tested on three-unit, four-unit, or long-span FDPs, regardless of the type of materials [18,19,25,26]. Although the pontic design of FDPs is more critical with connector area testing in the literature, the fabrication of two-unit FDPs in this study aids in minimizing the factors that could influence the results. If there is a pontic design, the force distribution in the three-point universal testing may affect the fracture resistance load and the mode of fracture readings. Additionally, the retainers (crowns) were designed with exact tooth dimensions that follow the abutment preparation to ensure the connector area is in the middle of the metal model. To resemble a clinical scenario with triangular connector designs, avoiding an unequal force distribution on the connector area that could influence the fracture properties of 3D-printed resin material is our priority in this study.

A study by Harshitha et al. in 2013 [34] tested the stress distribution using finite element analysis. The study found that the maximum Von Misses stresses are concentrated at the connector area, especially in the cervical region [34]. The cylindrical connector design was less stressed than the triangular design [34]. All-ceramic FDPs framework strength depends on the connector dimension [3]. Hence, it is necessary to enhance the strength of 3D-printed resin FDPs by manufacturing an adequate connector dimension.

In addition, Rezaei et al. in 2014 found the failure probability decreased with the high width of the connector [25]. For all ceramic prostheses, the connector dimension is more important than the abutment core thickness, as found by Ambre et al. in 2013 [16].

In another study, Modi et al. found that nonrigid connectors produced less stress than rigid ones [33]. In this study, evaluating the connector dimension on fracture properties for a ridge connector is essential. According to Shillingburg et al., the nonrigid connector transfers the stress to the supporting structure rather than concentrating it in the connector area [22]. In addition, Junker et al. (2019) recommended that minimum connector dimensions prove crucial for the load-bearing capability of monolithic e.max CAD FPDs [24]. Significantly, the reduction of connector dimension increased the strength of 3D-printed resin FDPs. According to the results, the mean of Modulus, Yield Strength, and Compressive Strength of 2 × 3 mm connectors were significantly larger among the other groups. In triangular connector shape design, the cross-sectional connector area decreases as the connector dimensions decrease, Table 1, which is the denominator that the fracture load will be divided on to produce the stress values in megapascals. We should remember that usually, in three-point bends of specimens that have relatively comparable cross-sections, stress can be compared easily since the denominator is almost equal. However, in this current study, we are comparing loads, moduli, and strengths that are different in cross-section, which makes it look odd that the fracture strengths were reduced as the cross-section decreased. Another way to look at it is to normalize those loads and strengths by unifying the denominators across the tested groups. Then, the results may look more relevant and comparable.

Larsson et al. tested all ceramic zirconia-based FDP connector diameters [26]. The authors evaluated the fracture strength properties of varied materials. This in vitro study recommended a minimum connector diameter of 4.0 mm for long spans or replacing molar teeth [26]. Compared to 3D-printed resin material, all ceramic material is harder than 3D-printed resin [35]. Hence, the thinning of the external wall around the large connector of 3D-printed resin FDPs could decrease the strength resistance load.

Observation of the failure mode on those dies shows almost similar fractures on both sides of the connectors and the crowns, yet no other fractures passed through the connectors themselves. This seems logical because fracture occurs in the weakest area of the designed restoration [19,36]. The highest number of fractures, counts, and percentages were in the 3 × 4 mm connectors compared to the other connectors dimensions. Regarding the location of the fractures and the fracture lines, it was observed that they are mostly on the occlusal and distal parts of the two adjacent crowns.

This in vitro study provides valuable insight, as 3D-printed resin is a relatively new material for provisional prostheses. Knowing its influence on fracture properties may help improve the design and printing process of these prostheses.

The manufacturing of each clinical or laboratory step must be precise or have a guideline for any new technology, which is why further investigation is needed. However, each material has an indication depending on its mechanical and physical properties and availability of either provisional or permanent prosthesis.

This study is in vitro, which may have a different outcome than the patient’s mouth and has some limitations. One limitation is that the specimens were tested statically using a universal testing machine. In addition, the tested provisional material and 3D printer are from one company. Adding another 3D-printed material and different connector-shaped designs will be helpful in the future. In addition, the specimens should be tested in a dynamic testing machine to simulate patient function.

5. Conclusions

The study found that the maximum load causing fractures in 3D-printed provisional material connectors was consistent, regardless of connector cross-section variations. Elastic modulus, yield strength, and compressive strength differed between 4 × 4 mm and 2 × 3 mm connectors, with the smallest showing the highest values. More connector fractures correlated with lower mechanical performance. The 2 × 3 mm group performed best, while the 4 × 4 mm group performed worst. Lower fracture counts, associated with single-direction fractures, indicated higher mechanical strengths, whereas multiple-direction fractures were linked to lower strengths.