The hemimandibles were removed with care to preserve the integrity of both peri-implant hard and soft tissues, washed in saline solution and fixed in 10% buffered formalin, and sent for processing at the Laboratory of Ucam-Biotecnos (Murcia, Spain). Specimens were dehydrated in ascending series of alcohol rinses and embedded in a glycol methacrylate resin (Technovit 7200 VLC; Kulzer, Wehrheim, Germany). After polymerization, the specimens were sectioned along its longitudinal axis with a high-precision diamond disk in the IsoMet® 1000 (Buehler, Lake Bluff, IL, USA), at about 150 μm down to 30 μm. A total of two slides were obtained for each implant. The slides were stained with Picrosirius Red Stain (Polysciences, Inc., Warrington, USA) and observed in a normal transmitted light microscope and a polarized light microscope (Nikon, Tokyo, Japan). Buccal bone wall level in comparison with lingual bone wall height after remodeling was expressed as a linear measurement from the implant shoulder to the first bone-implant contact, showed in the Fig. 2 corresponding with the A-B distance. The buccal and lingual tissue thickness was measured in the level corresponding with the implant shoulder (A line) from the implant to the external epithelium portion of the mucosa, showed in the Fig. 2 corresponding with the C-D distance. The measurements were performed by an expert examiner in histology (SG).

Metric evaluation of the predetermined parameters was carried out using a light microscope (Nikon, Tokyo, Japan) connected to a high-resolution video camera (3CCD, JVC KY-F55B, JVC®; JVC, Yokohama, Japan). After digitizing the phase of each specimen under the light microscope, all proposed details were measured in the images using the program Image Tool version 5.02 for Microsoft Windows™ (UT Health Science Center School of Dentistry, San Antonio, TX, USA).

Means, medians, and standard deviations of crestal bone height and tissue thickness were calculated for all groups. All data sets (n = 10) were tested for normality using the Shapiro–Wilk test, and the data did not show normal distribution. The Friedman test was performed for intergroup comparisons in buccal or lingual recorded measures for A-B and C-D parameters followed by the Dunn’s multiple comparison test for further comparison of different groups. Furthermore, Wilcoxon matched-pairs signed-rank test was used for the comparison of two groups. The significance level was set at P < 0.05.