Original Work
Relationship between bone-type alkaline phosphatase levels in gingival crevicular fluid and clinical parameters during supportive periodontal therapy
2018 Volume 60 Issue 1 Pages 26-34
Details
2018 Volume 60 Issue 1 Pages 26-34
Periodontal tissues such as periodontal ligaments are known to have high alkaline phosphatase (ALP) activity, and the genotype of ALP in periodontal ligament is reported to the bone-type. Therefore, if periodontal tissues are damaged by periodontitis, bone-type ALP (BAP) may be released in gingival crevicular fluid (GCF). The aim of this study was to compare the values of BAP in the GCF in patients during supportive periodontal therapy (SPT) with clinical parameters, as a contribution to knowledge regarding early clinical diagnosis of the disease. Thus, we sampled GCF from healthy sites (probing depth (PD) ≤4 mm and bleeding on probing (BOP) (-) ) and diseased sites (PD ≥4 mm and BOP (+) ) from 76 patients receiving SPT. We then measured clinical parameters; plaque index (PI), amount of GCF, gingival index (GI), PD, clinical attachment level (CAL), BOP and alveolar bone resorption. Biochemical parameters included the amount of BAP, aspartate aminotransferase (AST) activity and amount of protein. In addition, we defined a cut-off value for the amount of BAP for analysis.
The results demonstrated that diseased sites were significantly higher for all clinical and biochemical parameters than healthy sites. Furthermore, in an analysis using the cut-off values, those above the cut-off value in healthy sites had significantly higher values of AST activity and amount of protein when compared those with less than the cut-off value. Therefore, the amount of BAP may be related to condition of periodontal health. Based on this study, the amount of BAP in GCF may be indicative of early periodontal tissue damage.
In the clinical evaluation of periodontal disease, probing depth (PD) and bleeding on probing (BOP) evaluate the condition around the periodontal tissue, together with X-ray imaging1). However, the positive predictive value of BOP is reported to be low2), and some reports have noted inconsistencies with other clinical criteria3). By analyzing the components of gingival crevicular fluid (GCF), clinicians are able to evaluate the condition of periodontal tissues at each site, and it has been reported that GCF may contribute to early diagnosis of periodontal disease4).
To date, enzymes such as aspartate aminotransferase (AST), protein, and cytokines have been considered to be host-derived diagnostic markers observed in GCF, along with PD and BOP, to diagnose periodontitis4). Measurement of these inflammation markers is considered to be effective in determining the presence of inflammation, in treatment planning and whether any changes in treatment are needed for maintenance, but some reports do not agree that GCF analysis is a useful periodontal diagnostic tool5). Therefore, we aimed to improve diagnostic accuracy for periodontitis by analyzing inflammatory and bone metabolism markers.
Periodontitis is an inflammatory disease that causes connective tissue and bone degradation, which can result in the tooth loss1). Therefore, in this study, we focused on the markers associated with bone metabolism and particularly examined alkaline phosphatase (ALP). ALP is an enzyme associated with bone formation6) and it is known to be widely distributed throughout the body. ALP is anchored to glycosyl-phosphatidylinositol, mostly on cell membrane, and it is released into serum in response to tissue degradation, such as through inflammation7,8). ALP activity is especially high in GCF in periodontitis, and it has been reported that there is high correlation between ALP activity and clinical conditions in periodontitis9).
ALP is classified into four types of isozyme, each with different glycoprotein genes. These include tissue nonspecific type (bone/liver/kidney-type), intestinal type, placental-type and placental-like type10-12). Bone-type ALP (BAP) in this classification is mainly possessed by osteoblasts, and it decomposes the inorganic pyrophosphoric acid that suppresses calcification and is involved in bone formation13). From these reports, an enzyme immunoassay (EIA) kit that utilized the monoclonal antibody against BAP was developed, and is presently used in clinical examinations to assess the state of osteoblasts and bone formation. Furthermore, it has been reported that BAP is increased in the blood following bone loss due to bone remodeling and is decreased in areas of high calcification during frequent bone remodeling. Measurement of BAP is performed in the diagnosis of bone metabolism diseases, such as bone tumors14), Paget's disease of bone15), hyperthyroidism16) and osteoporosis17). Furthermore, periodontal ligament cells have an extremely high level of ALP activity18). The type of ALP in such cells is reported to be BAP18-20). Goseki et al., using classical methods, reported that BAP is present in GCF21), while Isik et al.22) reported that the amount of BAP in the GCF changes following teeth movement in orthodontic treatment. In addition, Otes et al.23) sampled GCF from Cynomolgus monkeys with experimental periodontitis and measured the amount of BAP. Based on the above reports, release of BAP occurs when periodontal tissues are damaged by periodontitis. However, there have been no reports on the relationship between periodontal disease and the amount of BAP in GCF. The present study aimed to compare the values of BAP in GCF from patients during supportive periodontal therapy (SPT) with other clinical parameters, in order to contribute to knowledge on early clinical diagnosis of the disease.
Subjects were 76 non-smokers (39 men, 37 women; age, 63.4±10.8 years) who were receiving SPT at Nippon Dental University Hospital. Inclusion criteria were that the patient had at least 12 teeth remaining and were generally healthy. Exclusion criteria were as follows: 1) patients with diabetes, immune disorders, liver disease, heart disease or osteoporosis; 2) females who were pregnant or are taking birth control pills; 3) patients who had received antimicrobial therapy for the past 3 months; and 4) patients who did not provide consent to this study. Prior to commencing this study, subjects received an explanation of the study and were asked to provide written consent. This study was conducted in accordance with the Helsinki declaration, with approval from the ethics committee of Nippon Dental University Hospital (approval number, NDU-T 2014-56).
Demographic data of the patient population
GCF was sampled from a single tooth for each of the following clinical parameters from the diseased and healthy sites of the same patient. Criteria for sampling sites were as follows: for healthy sites, PD ≤4 mm and BOP (-); for diseased sites, PD ≥4 mm and BOP (+). Abutment teeth for partial dentures and full crowns were excluded. Each of the clinical parameters were assessed and recorded by a specialist periodontist (HI) in the following order: PlI24), amount of GCF (Periotoron® 8000; Oraflow, Plainvew, NY), PD, CAL, GI25), BOP and alveolar bone resorption. For measurement of alveolar bone resorption, we used the modified ruler by Olav et al.26); dental x-rays were taken from each sites and alveolar bone resorption ratio was measured from the root length and alveolar bone height.
The Williams probe (Hu-Friedy Inc., Chicago, IL) was used for measurement of PD, CAL and BOP. To sample GCF, after measuring the PlI, the area was simply dried using cotton rolls and air, and the overlying plaque was removed as much as possible. Subsequently, PerioPaper® (Oraflow) was inserted until resistance was felt in the pocket, and after 30 seconds, GCF was sampled. The process for collecting GCF was repeated three times. The amount of sampled GCF (μl) was measured by preparing a calibration curve with Periotoron®. The PerioPaper® used to sample the GCF was immediately soaked in saline, and after five minutes of stirring, was centrifuged for five minutes at 10,000 rpm. The supernatant was dispensed for biochemical analysis. The samples for which blood was found on the PerioPaper® were excluded, and each sample was stored at -80°C until analysis was performed.
Analysis of biochemical parametersBiochemical parameters included amount of BAP, AST activity and amount of protein. Amount of BAP was expressed as μU/pocket, applying the EIA method (Osteolinks™ BAP; DS Pharma Biomedica, Osaka, Japan). BAP was selectively captured with a mouse anti-BAP monoclonal antibody, and p-nitrophenyl phosphate was added. The enzyme reaction was performed with the trapped BAP, and the color after the enzyme reaction was measured using a microplate reader at 405 nm. AST activity was expressed as μU/pocket using the POP-TOOS method (Wako Pure Chemical Industries, Ltd., Tokyo, Japan), which expresses color through oxidative condensation by the action of pyruvate oxidase under dissolved oxygen. To measure AST activity, GCF was added to saline with 5% bovine serum albumin (BSA; Sigma-Aldrich, St. Louis, MO). To measure protein, we used the BCA™ Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA) and values are expressed as μg/pocket.
Statistical analysisClinical and biochemical parameters are expressed as means and standard deviation (SD). The analytical software was used SPSS ver. 22.0 J (IBM-SPSS, Inc., Chicago, IL). To compare healthy and diseased sites, we used the Wilcoxon signed-rank test. Correlations between clinical and biochemical parameters were examined using the Spearman correlation coefficient. In addition, based on the criteria for healthy and diseased sites, we prepared a cut-off value for the amount of BAP using the receiver operating characteristic curve and the Youden index. Below the cut-off value was defined as low BAP (L-BAP), and above the cut-off value was defined as high BAP (H-BAP). Furthermore, we created L-BAP and H-BAP for both healthy and diseased sites (four groups): healthy L-BAP sites; healthy H-BAP sites; diseased L-BAP sites; and diseased H-BAP sites. We performed Kruskal-Wallis test and Steel-Dwass test (post-hoc test) between each group for all clinical and biochemical parameters, and examined the usefulness of the amount of BAP for clinical application. Statistical significance was set at P<0.05.
The work flow of this study is shown in Fig. 1.
Flowchart describing the classification of 152 GCF samples from 76 patients
Healthy sites (n＝76) were defined as having a PD ≤4 mm and BOP (－). Diseased sites (n＝76) were defined as having a PD ≥4 mm and BOP (＋). We created L-BAP sites and H-BAP for healthy and diseased sites. Four groups: healthy L-BAP sites (n＝66), healthy H-BAP sites (n＝10), diseased L-BAP sites (n＝57), diseased H-BAP sites (n＝19).
Table 2 shows the values for the clinical and biochemical parameters. The findings show that the diseased sites were statistically significantly higher in all clinical and biochemical parameters than the healthy sites (P<0.01).
Parameter of sampling sites
There was a correlation between the amount of BAP and AST activity in all groups (P<0.01). Among the clinical parameters, a correlation was demonstrated between alveolar bone resorption and amount of BAP at diseased sites (P<0.05).
Correlation between amount of BAP and clinical parameters and biochemical data
The cut-off value for the amount of BAP was 374.35 (μU/pocket), and the sensitivity and specificity of BAP were 0.250 and 0.868, respectively. For L-BAP and H-BAP at healthy and diseased sites, the clinical and biochemical parameters were examined using the cut-off value. The results of Kruskal Wallis test showed a statistically significant difference in all clinical and biochemical parameters (P<0.01). Furthermore, for post-hoc analysis, we used the Steel-Dwass test. The results demonstrated that the healthy L-BAP and H-BAP sites did not show a statistically significant difference in any of the clinical parameters. However, when compared to healthy L-BAP sites, healthy H-BAP sites showed significantly higher values of AST activity and amount of protein (P<0.05). Diseased L-BAP and H-BAP sites did not show a significant difference in any of the clinical or biochemical parameters. Furthermore, when compared to the healthy H-BAP sites, clinical parameters (GI, PD, CAL, amount of GCF and alveolar bone resorption) showed significantly lower values when compared with diseased L-BAP sites. However, there were no significant differences in any biochemical parameters in comparison with diseased L-BAP sites as compared with healthy H-BAP sites (Table 4).
Comparison of the parameter of each group
The most important factor to consider during the SPT period of periodontal treatment is to accurately grasp the patient's symptoms, with the aim of early detection and providing early treatment4). Lang et al.2) noted that determining the rate of positive BOP during the maintenance period is extremely important in prognosis estimation. However, an examination that used cut-off values for the clinical parameters and enzyme activity in GCF found that, regardless of the severity of periodontal disease, there was disagreement in the BOP results and the limitations of conventional examinations for periodontal disease3,27). Therefore, additional periodontal examinations, such as compositional analysis of GCF, are now a focus of research4). To date, several host-derived diagnostic markers for periodontal disease in GCF have been reported, and ALP has been suggested as a useful marker that contributes to treatment planning and monitoring of periodontal tissue during the SPT period9,28,29). Ishikawa et al.30) reported a correlation between ALP activity, and PD and alveolar bone resorption. Similarly, Chapple et al.9) discussed a correlation between loss of attachment and ALP activity. Based on these reports, measurement of ALP activity in GCF has been recognized as an extremely useful host-derived diagnostic marker for periodontal disease in the field of periodontology, and several follow-up reports have been published31,32).
It was reported that ALP in GCF consisted of several ALP isoenzyme types whose possible origins are derived from phosphatidylinositol-anchored ALP and periodontopathic bacterial ALP21). In addition, periodontal ligament shows extremely high levels of ALP activity18), and the genetic type is reported to be bone-type18-20). Therefore, it was speculated that BAP associated with damage to periodontal ligament can be observed in GCF.
Thus, we compared the amount of BAP in GCF and clinical parameters for both healthy and diseased sites from SPT using the split-mouth design to exclude the possibility of individual differences due to different oral environments. Lamster et al.33) found that total volume of GCF be used to reflect the state of periodontal tissue, rather than concentration. Therefore, we analyzed total volume. Our results showed that clinical and biochemical parameters showed significantly higher values at diseased sites. In addition, BAP in GCF was assumed to be derived from the periodontal ligament, while ALP on the cell membrane surface is known to be separated by glycosylphosphatidylinositol phospholipase D (GPI-PLD), which is abundantly present in plasma isolate34). Fujikawa et al.35) reported that an increase in PD and GI leads to a shift to more acidic values of pH in the pocket. Furthermore, Eggert et al.36) reported that GCF at sites with inflamed periodontal tissue tends to have acidic pH. In other words, if there is inflammation in the periodontal tissue, the environment inside of pockets tends to become acidic, which is a favorable environment for GPI-PLD activity, leading to the release of BAP37). However, though a correlation between the amount of BAP and alveolar bone resorption at diseased sites has been confirmed, there was no correlation between other clinical parameters. This indicates that measurement of the amount of BAP in GCF is specifically useful in the evaluation of alveolar bone resorption. On the other hand, among the biochemical parameters, only AST activity showed a correlation with amount of BAP in both healthy and diseased sites, and clinical parameters such as GCF volume and PD were no different between the healthy L-BAP and H-BAP sites. Nonetheless, inflammatory markers such as AST activity and amount of protein showed higher values at healthy H-BAP sites than at healthy L-BAP sites. This likely indicates that BAP measurement can detect minor tissue damage that would not show clinical symptoms.
Presently among studies on the amount of BAP observed in GCF, Oates et al.23) used Cynomolgus monkeys to measure BAP by inducing inflammation of the periodontal tissue through ligation of teeth, and the results show that the amount of BAP in GCF of ligated teeth was elevated. Changes in the amount of BAP measured in human GCF have been reported following the movement of teeth due to orthodontic dentistry alone22). These results indicate that the amount of BAP, measured over 28 days of orthodontic treatment, decreased to a similar degree as other bone formation markers. However, as there was no information on the condition of periodontal tissues during orthodontic treatment period, the correlation between the amount of BAP and periodontal tissue condition due to orthodontic treatment the problem still remains. Conversely, a decrease in the amount of BAP could be clinically applied to detect ankylosis showing degeneration of the periodontal ligament. In ankylosis, the lack of buffering action due to degeneration of the periodontal ligament can lead to root fracture38), and BAP measurement may be used as an index to prevent root fracture. In addition, BAP measurement could be applied as a diagnostic indicator for dental implants. The fundamental goal of dental implant treatment is osseointegration, in which titanium and alveolar bone are directly integrated39). In other words, healthy peri-implant tissue may exhibit lower amounts of BAP than the peri-implantitis site, and BAP may therefore be an indicator of implant maintenance.
In the present study, significant release of BAP in GCF was observed at diseased sites when compared to healthy sites. On the other hand, there were no correlations between the amount of BAP and clinical symptoms at diseased sites. However, even at clinically healthy sites, biochemical changes were observed using cut-off values for BAP. Therefore, amount of BAP may be related to periodontal health. In the future, to clarify the causal relationship between BAP and periodontal disease, it will be necessary to conduct long-term observational research to estimate periodontitis activity.
The amount of BAP in GCF increased in the deep periodontal pocket with BOP (+) during SPT. Furthermore, BAP measurement using cut-off values confirmed that it is possible to detect biochemical changes even at clinically healthy sites. Therefore, measurement of BAP in GCF may help to assess early periodontal tissue damage.
This study was supported by a Grant-in-Aid from the Ministry of Education and Science Research Funds (Grant No. 17K11995・17K11996), and a 'New medical equipment and technology industry vision' project by the Japan Dental Association. We thank Syun-ichi Abe (Yamate Information Processing Center Ltd.) for advice regarding statistical analysis.
Conflict of Interest: None.
