Six females and 2 males (average age 66.4 years, range 52–85 years) with upper (n = 7) or lower (n = 1) complete dentures participated in this study. Inclusion criteria were (1) an opposite jaw that included natural dentition at least to the second premolar on both sides, (2) a need for fixed rehabilitation, (3) no medical contraindication to the placement of implants, (4) no need for augmentation procedures, and (5) willingness to participate in this study. The only exclusion criterion was temporo-mandibular dysfunction, since it may interfere with chewing and biting patterns and abilities. In the mandible 5 and in the maxilla, 6 OsseoSpeed implants and Uni Abutments 20° (Astra Tech, Mölndal, Sweden) were used to provide support for fixed rehabilitation. All participants were treated at the Center for Periodontology and Implantology, Leuven, Belgium, by the same surgeon (BC). Informed written consent with regard to treatment and masticatory function and follow-up procedures was provided to each participant. The study was approved by the ethics committee of the Catholic University of Leuven (B322201319432).
The day after implant surgery, implants were loaded with screw-retained implant-supported acrylic resin provisional restoration (immediate loading) as previously described [37, 38]. All provisional bridges extended to the second premolar or first molar region.
Occlusal contact area and approximate maximum bite force measurements
Patient’s head was positioned with the Frankfort plane parallel to the floor. After opening the mouth, a pressure-sensitive sheet (Dental Prescale, 50H, type R, 97 μm thick, GC, Tokyo, Japan) was inserted on the occlusal plane. Patients were instructed to bite onto the test sheet as hard as possible for 3 s in the intercuspal position. This was repeated three times in each patient. The sheets were analyzed using special analytical equipment (Occluzer FPD-707, GC, Tokyo, Japan), namely, an analyzing device that could calculate bite force (N) and occlusal contact area (mm2) from the degree of discoloration of the pressure-sensitive sheets. Values from three sheets were averaged for each measurement, as described in a previous study [39]. In a pilot study, dentate patients (n = 14, mean age 58.4 ± 12.6 years) showed an occlusal contact area of 20.79 ± 8.10 mm2 and a maximal bite force of 696.8 ± 237.5 N.
Measurement of masticatory efficiency
To assess the masticatory efficiency, we used glucose extraction in the filtrate obtained after chewing the specimen. After rinsing the mouth with tap water, a gum-like specimen mixed with 5% glucose with a height of 10 mm (Glucosensor Gummy, GC, Tokyo, Japan) was placed on patient’s tongue with chopsticks. Patients were requested to chew on the cube for 20 s, after which, they expectorated all the chunks of the cube into a cup equipped with a mesh filter to hold the debris. Thereafter, they rinsed their mouth again with 10 ml of water and expectorated into the same cup. The amount of glucose extraction in the filtrate obtained after chewing the specimen was used as a measure of masticatory efficiency. Glucose concentration in the filtrate (mg/dl) was measured using a calibrated Glucose Sensor Set (Glucosensor GS-1, GC, Tokyo, Japan), which utilizes a glucose sensor for diabetics (Accu-check Comfort, Roche Diagnostic, Basel, Switzerland) to measure masticatory efficiency according to a previous study, which reported its reliability for the evaluation of masticatory function [40]. For reproducibility, we tested the glucose concentration of control glucose solutions (500, 250, 125, 100, and 50 mg/dl) with the glucose sensor. The linear relationship that was observed between the glucose density of the solution (x) and the masticatory efficiency (the value of the glucose sensor) (y) is displayed in a scatter diagram (Fig. 1). The linear regression equation and Pearson’s correlation coefficient were as follows: y = 0.599 + 1.066x, r = 0.99 (n = 50, p< 0.0001). The intra-class correlation coefficient (ICC) is a prominent statistic to measure the test-retest reliability of data. The ICC (1, 3) of the data by Glucosensor was p = 1.000 (n = 5).
Food hardness assessment
Three types of chewing specimen with different levels of hardness (hard, medium, and soft), with the same size and taste, were produced from sucrose (800 g), glucose (870 g), sorbitol (1000 g), gelatin (hard, 390 g; medium, 240 g; and soft, 150 g), Arabia gum (hard, 36 g), citric acid (42 g), lemon juice (15 g), and water and were 15 × 15 ×10 mm in size. The hardness of each type was determined under maximal stress during compression of 9 mm with a crosshead speed of 100 mm/min with a tooth-shaped jig using a texture analyzer (EZ test, Shimadzu Co., Kyoto, Japan). The hardness results were 73 ± 1.5 N for the soft, 88 ± 1.5 N for the medium, and 171 ± 1.9 N for the hard specimens.
To assess the hardness differences, the examiner placed each test specimen on the tongue with chopsticks, and then the participants chewed on all sides and swallowed. They were asked to remember the hardness of the first specimen, which always had medium hardness and served as a control, and then to determine the level of hardness (hard, medium, or soft) of four consecutive and randomly administered specimens by comparing them with the first one. This test was conducted in a double-blind manner to eliminate examiner bias.
The number of correct answers of hardness was used as a measure of hardness recognition. The subjects were allowed to expectorate any specimen that could not be chewed well enough to be swallowed and could change their answers until the last specimen was chewed.
Data collection
Occlusal contact area, maximum bite force measurements, masticatory efficiency, and discriminating hardness assessments were performed on four occasions: (1) before implant surgery with the complete denture in situ, (2) 3 h after surgery, (3) 1–2 weeks, and (4) 3 months after insertion of the provisional screw-retained restoration.
Statistical analysis
Considering the small sample size in the present psychophysical experiments, the option was taken to report mainly the descriptive statistics, in terms of average (SD, range) values for bite force, occlusal contact area, glucose concentration, and number of correct answers regarding hardness. Some nonparametric analyses were added in the difference between baseline prior to surgery and the follow-up data (Wilcoxon test, SPSS for Macintosh ver.21, SPSS, Chicago, USA). A p value <0.05 was considered to be statistically significant.