Nine maxillary edentulous patients (seven males, two females) with a mean age of 68.0 ± 2.7 years (range 55–81 years) were recruited and voluntarily participated in this crossover study, which was conducted at Kyushu Dental University Hospital’s implant dentistry department. The implant surgery and fabrication of the prostheses were performed in accordance with the All-on-4™ treatment concept [1], in which a screw-retained provisional prosthesis is seated on four straight or angulated (17° and 30°) multiunit abutments (Nobel Biocare, Zürich-Flughafen, Switzerland) that are connected to the implants immediately after surgery. Computer-aided design/computer-aided manufacturing was used to fabricate the prostheses using a titanium-milled frame bonded with composite resin. The fitting surface of the prosthesis was fabricated in an ovate shape with contacts intimate to the residual alveolar ridge. After placement of the final prostheses, patients were verbally instructed to use their own manual toothbrushes to clean the boundary between the prosthesis and the alveolar ridge three times per day. As a routine follow-up cleaning protocol, each patient was recalled every 3 months, and the prosthesis was removed and cleaned with a toothbrush (Tuft24®MS, OralCare Inc., Tokyo, Japan) under running water and sterilized in an ultrasonic bath filled with chlorhexidine 0.05% (0.05 w/v MASKIN®/water, Maruishi Pharmaceutical Co., Ltd., Osaka, Japan) for 5 min. The study started 3 months after the final follow-up, at which point each participant had worn the final prosthesis for at least 1 year.
We investigated two electric-powered brushes (Sonicare Diamond Clean® attached to an HX6074/05 brushing head, Koninklijke Philips N.V., Amsterdam, the Netherlands [SD group], and the Oral-B Professional Care Smart Series 5000® attached to an EB20 brushing head, Braun GmbH, Kronberg, Germany [OralB group]) and one electric dental floss unit (Air Floss® attached to an HX8002/05 nozzle, Koninklijke Philips N.V. [AF group]). A manual toothbrush (Tuft24® MS, OralCare Inc., Tokyo, Japan) was used for the control group. Dentifrice was not used during the assessments in order to solely evaluate the ability of the instruments. To evaluate the efficacy of each brushing instrument, three electric instruments were randomly assigned for use along with 2-week washout periods between evaluations; the manual toothbrush was used last and served as the control. In this crossover study design, each instrument was used one time for 5 min during the respective assessment, and each participant used all four cleaning instruments (Fig. 1a, b). All participants confirmed that they had not been using any electric tooth cleaning instruments; thus, they used their own manual toothbrushes during the washout periods. The randomization sequence for the three electric instruments was created using Excel 2011 (Microsoft, Redmond, WA, USA). The study procedure was approved by the Kyushu Dental University Ethics Committee (approval number 13-3) and followed the guidelines of the amended Declaration of Helsinki. All participants provided written informed consent.
First, the fixed maxillary prosthesis was removed, and the plaque that had accumulated on the fitting surface was dyed with plaque-disclosing solution (Dent. liquid plaque tester, Lion Corp., Tokyo, Japan). Photographs of the dyed fitting surface were taken with a digital camera (D5200, Nikon Corp., Tokyo, Japan) (Fig. 1c). After the prosthesis was placed back in the mouth and retained by 15-Ncm screws, the patient was instructed to brush for 5 min in front of a mirror. Subsequent to removal of the prosthesis, a photograph of the fitting surface was taken to evaluate the remaining plaque. Finally, the remaining plaque was removed with a manual toothbrush (Tuft24® MS) under running water, and the prosthesis was replaced using 15-Ncm screws. Patients were only given basic guidance on use of the electric instrument, such as how to turn it on and off.
To calculate the dyed plaque area, tracing paper was mounted on the photograph. The margins of the prostheses and the plaque areas were traced manually. Only one individual performed the tracing; this individual was blinded to assignment, in order to ensure consistent assessment. The encircled plaque area was filled with red color (Fig. 1d) using an open-source image editor (GUN Image Manipulation Program) [16], followed by identification and calculation of the colored area with an open-source image-processing program (ImageJ) [17]. To determine the percentage of the area covered with plaque within the fitting surface of the prosthesis, the sum of the plaque area was divided by the entire outlined area of the fitting surface. To compare the plaque removal rates among the four groups, plaque-covered areas were directly compared before and after cleaning. Participants were then divided into two groups based on the plaque removal rate using a manual toothbrush; those with plaque removal rates ≤ 60% were assigned to the poor brushing group, and those with rates > 60% were assigned to the good brushing group. The buccal and palatal areas of the prosthesis were defined by dividing the implant area at the center line that passed through each of the four implants’ midpoints (Fig. 1e).
Kruskal-Wallis one-way analysis of variance, followed by Dunn’s multiple comparisons test, was used to determine statistically significant differences. Unpaired t tests with Welch’s correction were performed to compare the two groups. Statistical software (Prism7 Software, GraphPad Software, La Jolla, CA, USA) was used for all analyses; p < 0.05 was considered statistically significant. All results are presented as mean ± standard error. Statistical power analysis was performed using an open-source power analysis program (G*Power 3.1) [18].