Introduction

As indicated by Wolff’s law [1], bone is remodeled to meet its mechanical demands, suggesting that mechanical forces are among the most potent factors that influence bone formation and resorption. In contrast, the structure and volume of the bone are maintained by the tireless bone metabolism so that bone deformation is considered to be a consequence of the bone metabolic changes. Therefore, it is considered that the mechanical load induces metabolic changes in bone at the beginning as a mechanobiological reaction.

Excessive pressure due to wearing mal-adapting dentures is well known to cause bone resorption beneath the denture [2]. X-rays have been commonly utilized to evaluate the changes in the bone beneath the denture. However, X-ray images merely detect bone density and relatively large changes in the bone shape and structure, whereas a nuclear-based procedure can detect functional changes in the bone that occur prior to structural changes. In particular, 18F-fluoride positron emission computerized-tomography (PET) has received much attention in detecting bone diseases such as bone metastases because of its high image quality [3–5]. The functional imaging technique of 18F-fluoride PET allows a quantitative assessment of bone metabolism at specific sites of the skeleton in man [6–16], suggesting that the technique could be used to quantify the magnitude of the localized response to pressure induced by wearing dentures. In addition, 18F-fluoride PET should be useful for serial scanning to document the time course of the skeletal response after loading, because of the short half-life of 18F-fluoride (<2 h).

This study aimed to describe the time course of the bone metabolism at the residual ridge beneath the denture following removable partial denture (RPD) use by 18F-fluoride PET/computed tomography (CT) scanning.

Materials and methods

Subjects

A 66-year-old woman (subject A), a 67-year-old woman (subject B), and a 65-year-old woman (subject C) without abnormal bone metabolism consulted the Advanced Prosthetic Dentistry of Tohoku University Hospital. Their chief complaint was masticatory dysfunction on the edentulous region of the dental arch. Subject A had lost her mandibular left molars more than 8 years, subject B had lost her mandibular left molars and second premolar more than 1 year, and subject C had lost her mandibular right molars more than 1 year before presenting to our hospital. They had no experience with mandibular dentures (Fig. 1a). All three subjects chose unilateral distal extension RPD replacing the edentulous region of the dental arch (Fig. 1b). We fabricated a platinum metal base RPD for subject A, a resin base RPD for subject B, and a cobalt chrome metal base RPD for subject C (Fig. 1c). The adaptation of the denture base to the residual ridge was checked using white silicon (Fit checker, GC Co., Tokyo, Japan) at baseline and 13 weeks after RPD use. Occlusal surfaces of the artificial teeth were adjusted to distribute the symmetrical occlusal contacts within the dental arch. A day after the first use of the RPD, a postinsertion adjustment was made.

Fig. 1
figure 1

Subjects and definition of VOI and ROI. a Occlusal views of the mandibles. Dashed line indicates edentulous region of dental arch; b occlusal views of the mandibles with the RPDs replacing edentulous regions indicated by dashed lines; c RPDs; d definition of VOI; e definition of transaxial plane number. ROI: edentulous regions of mandibles in transaxial planes

Full size image

18F-fluoride PET/CT imaging

The metabolic changes in the residual bone beneath the RPD were assessed using an 18F-fluoride PET/CT imaging scanner (Discovery ST Elite, GE Healthcare Japan Co., Tokyo, Japan). 18F-fluoride PET was performed at baseline, and 6 and 13 weeks after RPD use for subjects A and C, and at baseline, and 4 and 13 weeks for subject B. Research protocols for this study were approved by the research ethics committee at both the Tohoku University Graduate School of Dentistry and Sendai Kousei Hospital. Signed consent forms were obtained after full explanation of the procedures from the three subjects. The emission scans in the three-dimensional acquisition mode with spatial resolutions of 2.0, 2.0, and 3.27 mm in the radial, tangential, and axial directions (Table 1) were started 75 min after intravenous injection of 37 MBq 18F-fluoride. Subjects were positioned supine with the jaw bone in the field of view (FOV). The scan time of the jaw bone was extended from the regular 3 to 20 min to obtain clear PET images even with a low 18F-fluoride dosage (1/5 the regular dosage per scan). In all cases, low-dose CT was performed with exposure parameters of 120 kV, tube current 15 mA, and 3.75 mm slice thickness.

Table 1 Scanning condition
Full size table

Data analyses

The PET and the CT data were processed and fused using medical image viewer software (EV Insite R, PSP Co., Tokyo, Japan) to identify the anatomical location. Subsequently, the volumes of interest (VOIs) were placed on their mandibles at the edentulous regions beneath the RPDs, which included the mandibular bone from the distal end of the direct abutment to the distal border of the retromolar pad (Fig. 1d). Furthermore, the regions of interest (ROIs) were placed at the edentulous regions of the mandibles beneath the RPDs in the transaxial planes numbered from superior, 1, to inferior, 11 (Fig. 1e). CT values and the mean standardized uptake values (SUVs) were measured for each ROI. Subjects A, B, and C had 7, 11, and 9 ROIs, respectively. The SUV of VOI, defined as the average value of SUV of all ROIs in each subject, was calculated. The CT value of VOI was also calculated in the same way. In each subject, the differences between the serial CT values of VOI and between the serial SUVs of VOI were tested using two-way analysis of variance (ANOVA) and Dunnett test for post hoc analyses (Dr. SPSS II for Windows, SPSS Inc., Chicago, IL, USA). The SUV of ROI was analyzed to examine the effect of time and distance from the RPD.

Results

The adaptation of the denture base to the residual ridge was successful at both baseline and 13 weeks after RPD use. An interview and questionnaire revealed that the masticatory function was improved after RPD use and that there was no trouble such as pain at the residual ridge beneath the denture in any of the three subjects.

The SUVs of each VOI were significantly increased at 4–6 weeks after RPD use and then decreased at 13 weeks in all three subjects (P < 0.05; two-way ANOVA, Dunnett test) (Figs. 2, 3). The CT images, on the other hand, showed no obvious changes in the bone shape or structure beneath the RPDs, and the CT values of each VOI remained static after the RPD use in all three subjects (Fig. 3). Figure 4 showed serial changes in the SUVs of each ROI during treatment with RPDs. Except for transaxial plane 1, the SUVs of ROIs tended to increase as the number of the transaxial planes decreased, or in other words, as the distance from the RPD decreased. This tendency became pronounced after RPD use. Moreover, the SUV of ROIs in transaxial planes 2–8 tended to increase after RPD use, whereas those in planes 9–11 remained static.

Fig. 2
figure 2

Transaxial plane images, indicated by a line in left diagram, of accumulation of 18F-fluoride at baseline, and 4 and 13 weeks after RPD use. Arrow indicates region of residual ridge beneath the RPD

Full size image
Fig. 3
figure 3

Percentage changes in the SUVs and the CT values of VOIs. *P < 0.05; two-way ANOVA, Dunnett test

Full size image
Fig. 4
figure 4

Changes in the SUVs of ROIs

Full size image

Discussion

Radiographic methods have been predominantly used not only in clinical practice but also in human studies of residual ridge resorption. In these studies, X-ray images were usually taken before denture use and a few years afterwards [17–25]. In our previous report, we established a scanning procedure to detect bone metabolic changes beneath the RPD using 18F-fluoride PET/CT [26]. Clear PET images were successfully obtained with low-level radiation exposure by extending the scan time to 20 min, which enables us to scan several times per subject. In this study, we applied this procedure to three subjects and examined the detailed dynamic and longitudinal processes in the biologic responses of residual bone beneath the denture immediately after the first RPD use.

In the present investigation, the SUVs of VOIs and ROIs in transaxial planes 2–8 increased after RPD use. This may represent not increased blood flow as an inflammatory reaction but rather increased bone metabolism as a mechanobiological reaction to the pressure due to RPD use based on the following: (1) No pain or inflammation is reported in the denture supporting tissue in clinical practice. (2) If it is inflammatory reaction, the response of the periosteum, which is highly vascular, would be marked. Actually the SUVs in transaxial plane 1 including the periosteum were not so high. (3) In the case of inflammatory reaction, blood flow increases for fewer than 2–3 weeks, whereas the SUVs in the present cases increased at 4–6 weeks.

Our previous animal studies of bone metabolism beneath the denture base and around loaded implants using bone scintigraphy [27, 28] showed that relatively low mechanical stress initially increased bone metabolism to the peak level, which declined to the baseline level with time. In this study, the SUV changes revealed a similar trend to that in our previous studies, and the SUV initially increased at 4–6 weeks after RPD use and then decreased at 13 weeks in all three subjects. The adaptation of the denture base to the residual ridge was successful, and no pain or inflammation was reported in the denture supporting tissue, implying that this change may be temporary and may have been related to the process of bone adaptation to the RPD. Furthermore, our previous animal study also indicated that the change pattern of bone metabolism during 6 weeks after loading depended on the magnitude of the load [28]. Thus, excessive stress may produce a different SUV change pattern in human case studies during 13 weeks or so.

Low SUVs of ROIs in transaxial plane 1 were most likely due to the fact that transaxial plane 1 contained a high level of cortical bone whereas the other transaxial planes contained mostly cancellous bone. The absence of change in the SUVs of ROIs in transaxial planes 9–11 may be caused by mechanical loading due to RPD use not being transmitted to those planes, which were far away from the RPDs.

In conclusion, this study indicates that in first-time RPD users, wearing of a well-adapted RPD initially increases bone metabolism beneath the denture after which it decreases at around 13 weeks after RPD use without any bone structural changes detectable by clinical X-rays. This metabolic change is a mechanobiological reaction to the pressure induced by RPD use.