Sample size estimation

A power analysis was designed to have adequate power to apply a two-sided statistical test of the null hypothesis that there is no difference between the tested groups regarding fracture resistance. By adopting an alpha (α) level of 0.05 (5%), a beta (β) level of 0.2 (i.e., power = 80%), and an effect size (d) of 1.51 calculated based on the results of a previous study (Makawi and Khattab 2019), the predicted total sample size (n) was found to be 16 samples (i.e., 8 samples per group). Sample size calculation was performed using G*Power version 3.1.9.7. (Faul et al. 2007). The sample size was increased by (25%) 20 samples (i.e., 10 samples per group) to account for possible failures in testing.

Ethics approval

This study was approved by the Research Ethics Committee, Faculty of Dentistry (FDASU-REC). The ethical approval number was FDASU-RECEM122). It was an in vitro study that used teeth collected from anonymous patients.

Teeth selection

Thirty human carious primary molars were collected from anonymous patients from the Department of Pediatric Dentistry and Dental Public Health, Faculty of Dentistry. The teeth were extracted due to overretention or for orthodontic reasons. Only 20 molars were selected for the study according to the following criteria: free of developmental disorders, no obvious cracks, and no previous dental restorations. For standardization, teeth were selected with an average buccolingual dimension of 6.5 ± 1 mm and mesiodistal dimension of 9.5 ± 1 mm measured using a digital caliper. All soft tissue debris was removed with a hand scaler and then stored in distilled water until usage for a maximum of 1 month (Simsek and Derelioglu 2016).

Pulpotomy procedure

Caries was removed with a high-speed handpiece using carbide round bur (sizes 4 and 5) under copious irrigation. Then, the access cavity was prepared, and pulp tissue was removed using a large sharp spoon excavator (tip size 1 mm). Then, a thick mix of zinc oxide and eugenol paste was applied to seal the canal orifices, followed by a light cured resin reinforced glass ionomer restorative material (GC Fuji II ®, GC).

Sample allocation

All samples were allocated randomly by the simple random sampling method using the Research Randomizer software program (https://www.randomizer.org/). An independent person generated randomization codes in sequentially numbered, secured, opaque wrappers to ensure covert distribution into two groups (Mahfouz Omer et al. 2023). Group 1: 3D printed crowns group (n = 10) was restored by NextDent resin (NextDent C&B MFH, Soesterberg, The Netherlands). Group 2: zirconia crowns (ZC) group (n = 10)—restored by prefabricated zirconia crowns (NuSmile ZR [NS], NuSmile, Houston, TX, USA).

Group 1: 3D printed crownsTooth preparation

The teeth were prepared using high speed contra-angled handpiece with blue-coded tapered diamond stone with round end (Mani TR-12) for buccal, lingual, mesial, and distal walls for 0.7–1 mm, producing a chamfer margin circumferentially, and then reduction of the occlusal surface was done using a blue-coded wheel stone (Mani WR-13) to produce occlusal clearance of 1.5 mm. The samples were then scanned using an intraoral scanner (Cerec Omnicam, Sirona, Bensheim, Germany) (Ahmad et al. 2023) Fig. 1.

Fig. 1

Scanning of a prepared molar using an intraoral scanner

3D crown designing

3D printed resin crowns were designed using the Exocad software (Exocad Gmbh, Darmstadt, Germany) to have a uniform thickness on all surfaces (average 1 mm), including occlusal, buccal, lingual, and proximal surfaces. After reviewing each design, it was exported as a high-resolution STL file to be outsourced and 3D printed (Alshamrani et al. 2023).

MICRODENT 1 PRO printer was utilized to print the 3D printed crowns using NextDent C&B MFH resin via digital light processing (DLP) technology. After printing, the platform was removed from the 3D printer and placed on a paper towel with the printed crowns facing upward. The printed crowns were separated from the platform and rinsed twice in a 96% alcohol solution ultrasonic bath to get rid of any excess material. The duration of each rinse was approximately 3 min to avoid causing defects in the printed crowns. The second rinse was carried out with a clean 96% alcohol solution. After cleaning, the printed crowns were dried to ensure that they were free of alcohol residue (Çakmak et al. 2021).

Printed crowns were then placed in a UV-light curing box with wavelength 300–500 nm (DENTCURE 2 by ZYLO3D) for post-curing according to the manufacturer’s instructions. The printed crowns were finished and polished using a conventional rotary machine.

Each crown was fitted on its prepared molar and assessed according to clinical practice under visual inspection using a 3.5X dental loupes (Lupa Bioart 3.5X) and the dental explorer. The 3D printed crowns were then cemented onto the prepared molars according to the manufacturer’s instructions with resin cement (TOTALCEM).

Excess cement was removed with a sharp dental explorer, and then the samples were stored in distilled water at 37 °C for 24 h (Ahmad et al. 2023).

Group 2: Zirconia crownsTooth preparation

The teeth were prepared using high speed contra-angled handpiece with blue-coded tapered diamond stone with a round end (Mani TR-12) for buccal, lingual, mesial, and distal walls for 0.7–1.75 mm, and then reduction of the occlusal surface was performed using a blue-coded wheel stone (Mani WR-13) to produce occlusal clearance of 1.5–2 mm, taking into consideration that the preparation must be free from any undercuts to avoid crown fracture.

The proper size of the crowns was selected using the NUSMILE Zirconia Try-In Kit; then it was removed, and the same size of the crowns as that obtained from the kit was cemented using resin cement (TOTALCEM).

After cementation, excess cement was removed with a hand explorer and the samples were stored in distilled water at 37 °C for 24 h (Ahmad et al. 2023).

Laboratory stepsThermocycling

The samples were subjected to 5000 thermal cycles with distilled water at 5 °C–55 °C, with dwell time = 25 s and lag time = 10 s using Mechatronic (Germany).

Marginal gap measurements

The samples were examined at the Department of Oral Pathology, Faculty of Dentistry, using a stereomicroscope (LG-PS2,Olympus,Japan) with a digital camera (Canon EOS 650D, Japan) at 40× magnification (Al-Haj 2019). The marginal gap was determined using the criteria proposed by Holmes et al., who defined the vertical marginal gap as the distance between the crown margin to the edge of the finish line preparation (Refaie et al. 2023). For each sample, eight points along the margins for each axial surface were captured. Fig. 2. Then, linear measurements were carried out using image analysis software (Image J, 1.41a, NIH, USA). The average marginal gap for each sample was then calculated. The gap distance was measured in micrometers.

Fig. 2

Diagram showing the eight points along the margins for each axial surface of each sample

Fracture resistance measurementsPreparation of samples for fracture resistance test

Samples were embedded perpendicularly in polyvinyl chloride (PVC) cylinders with the occlusal plane parallel to the ground using self-cure acrylic resin starting 1 mm below the cemento-enamel junction. To simulate the periodontal ligaments, a single layer of Teflon tape was wrapped around the roots of the teeth before being embedded in the acrylic resin.

Each sample was individually mounted on a computer-controlled universal testing machine (Instron Universal Testing Machine, England) and then secured to the lower fixed compartment of the testing machine by tightening screws. An axial loading was applied at the center of the crown using a stainless steel round-ended load applicator (3.6 mm diameter) attached to the upper part of the universal testing machine at a crosshead speed of 1 mm/min (Bani-Hani et al. 2025). The load at failure was manifested by an audible crack and confirmed by a sharp drop in the load–deflection curve recorded in the computer software (Bluehill 3 Universal materials testing software). The load required to fracture was recorded in newton (N) (Beattie et al. 2011).

Statistical analysis

Numerical data are presented as mean and standard deviation values. They were checked for normality by viewing the distribution and the Shapiro–Wilk's test. Data were found to be normally distributed and analyzed using an independent t test. The significance level was set at P < 0.05 within all tests. Statistical analysis was performed with R statistical analysis software version 4.3.1 for Windows (Team RC 2023).