A control model standard was fabricated from a cast obtained from a completely edentulous mandible. Any existing undercuts were blocked. The duplication of the stone model was carried out using laboratory addition silicone material (REPLISIL 22N, dent-e-con, Germany) to obtain an epoxy resin model (Swiss Chem; construction chemicals, Egypt).
For surgical guide construction, Cone Beam Computed Tomography (CBCT) (parameters 85 KVP, 5 MA) was recorded for the model to create a DICOM file. The standard tessellation language file (STL file) of the model was also obtained using an intraoral scanner (MEDIT i700; MEDIT Corp). The STL file was superimposed on the DICOM file using the best-fit algorithm.
Using computer-aided design software (Exocad GMBH Dental CAD), a virtual lower denture was designed.
A prosthetically driven implant placement concept was clear in mind, therefore a surgical guide using the virtual designed lower denture was constructed by an implant planning software (real guide; 3diemme, Italy). The implants were placed bilaterally between the lateral incisor and canine. The surgical guide was printed using clear surgical guide resin (EPAX Clear Resin; EPAX 3D). The resultant surgical guide was finished and cleaned with alcohol to remove excess monomer.
Two dental implants (Internal Tapered; BIOHORIZONS) of 3.8 mm in diameter and 10.5 mm in length were loaded into the model. Two castable plastic abutments were tightened using a torque wrench at 25 N over the implants. With the aid of a dental surveyor, the plastic bar with a round cross-section was attached to the two plastic abutments using self-cure acrylic resin, creating a 2 mm clearance space between the bar and the model. The assembly was then unscrewed, cast in Ni–Cr alloy, and finished and polished. The finished bar was checked for passivity and then finally screwed to the implants (Fig. 1).
The model with the bar screwed on was scanned using a desktop scanner (Medit T500) after the application of scan spray powder (Alldent, Germany) to get the STL file of the model and bar. The STL file of the bar and model was imported to the Meshmixer software (MESHMIXER 3.5 software, Autodesk).
First, the PEEK clip design was drawn on the model by outlining the lingual, buccal, mesial, and distal extents (Fig. 2). The boundaries were smoothened. Undercuts were created on the buccal and lingual aspects of the bar clips by using an attract brush tool to ensure mechanical retention between the clip and the denture fitting surface in the pickup step [24] (Fig. 3).
The designed clip was printed using (EPAX Dental Castable Resin; EPAX 3D). The clip wax pattern was pressed by the lost wax technique into PEEK (Fig. 4).
The epoxy model was duplicated into the stone cast on which 24 mandibular trial denture bases with waxed-up acrylic resin teeth (Zhengzhou Linker Medical Equipment Co., Ltd.) were fabricated. Mandibular trial dentures were flasked and packed with heat-cured resin (Denture Base Material; Vertex-Dental B.V.), then finished and polished with a hock attached to the denture geometric center.
The tested groups were classified as follows:
A light body rubber base was loaded into the denture, and the denture was tried on the model with a bar and clips loaded on it. Any pressure areas preventing the denture from complete seating or any areas responsible for denture frictional retention were removed. The retromolar pads were used as a reference for the complete seating of the denture base on the model [18].
Teflon and wax material were used to block any undercuts beneath the bar. Escape holes were made on the denture to act as an exit for the extra pickup material. The pickup was done by self-cured acrylic resin with the denture seated completely over the model. (Figs. 7 and 8).
The specimen’s retention forces were measured using the Instron universal testing machine (model 3345; England). The denture was attached to the dynamic part of the universal testing machine via a screw hook. The direction of the pull forces was performed vertically.
The achieved maximum values of retention force were recorded at the beginning of the study (initial retention) and after 1, 2, and 3 years with an average of 1000 cycles per year based upon patients’ average of 3 insertions and removals per day [22, 25]. Twenty-four epoxy models and dentures were used (12 for each group) for proper statistical sample sizing.
The sample size was calculated using G Power version 3.1.9.2. and according to previous studies [8, 25, 26].
Statistical methodology
The data was collected and entered into the computer using the SPSS (Statistical Package for Social Science) program (version 21). The data was normally distributed by the Kolmogorov–Smirnov test of normality, so the parametric statistics were adopted. The mean, standard deviation, and 95% CI of the mean were used to describe the data.
Two studied independent, normally distributed variables were compared using an independent sample T-test. Repeated measures analysis of variance was used. Model assumptions were tested and found to be satisfactory except for Mauchly’s test of sphericity.