**3.1 Outline of the Niigata Study**

74 Chronic Kidney Disease

A wide range of oral mucosal lesions, particularly white patches and/or ulceration, have been described in individuals receiving dialysis and allografts (Proctor et al.,

Kaposi's sarcoma (KS) can occur in the mouths of immunosuppresed renal transplant recipients (Farge, 1993). Any increased risk of oral malignancy in CRF probably reflects the effects of iatrogenic immunosupression, which increases the risk of virallyassociated tumors, such as KS or non-Hodgkin's lymphoma (Proctor et al., 2004).

*Candidosis*, angular cheilitis has been described in up to 4% of hemodialysis and renal allograft recipients (King et al*.*, 1994; Klassen and Krasko, 2002). Other oral candidal lesions—such as pseudomembranous (1.9%), erythematous (3.8%), and chronic atrophic

*Viral infection*, prior to the availability of appropriate anti-viral drugs (*e.g.*, acyclovir, gancyclovir, and valacyclovir), about 50% of renal allograft recipients, who were seropositive for herpes simplex, experienced recurrent, severe, and prolonged HSV infections (Armstrong et al*.*, 1976). However, in recent years, the use of effective antiherpetic regimes has significantly reduced the frequency of such infection (Kletzmayr et

Delayed eruption of permanent teeth has been reported in children with CRF (Wolff et al*.*, 1985; Jaffe et al*.*, 1990). Enamel hypoplasia of the primary and permanent teeth (Kho et al*.*, 1999; Koch et al*.*, 1999; Al Nowaiser et al*.*, 2003) with or without brown

A wide range of bone anomalies can arise in CRF. These reflect a variety of defects of calcium metabolism including, loss of hydroxylation of 1-hydroxycholecalciferol to active vitamin D (1,25-dihydroxycholecalciferol), decreased hydrogen ion excretion (and resultant acidosis); hyperphosphatemia, hypocalcemia and resultant secondary hyperparathyroidism and interference with phosphate metabolism by dialysis (Nadimi

Orofacial features of renal osteodystrophy due to hyperparathyroidism include bone demineralization, decreased trabeculation, decreased thickness of cortical bone, ground-glass appearance of bone, metastatic soft-tissue calcifications, radiolucent fibrocystic lesions, radiolucent giant cell lesions, lytic areas of bone, jaw fracture (due to trauma or during surgery) and abnormal bone healing after extraction. Orofacial features of renal osteodystrophy related to tooth and periodontium include delayed eruption, enamel hypoplasia, loss of the lamina dura, widening of the periodontal ligament, severe periodontal destruction, tooth mobility, drifting, pulp calcification and pulp narrowing (Damm et al*.*, 1997; Okada et al*.*, 2000; Klassen and Krasko, 2002).

**2. The relationships among osteoporosis, renal function and periodontal** 

Osteoporosis is the most common metabolic bone disease among the elderly, and the incidence of osteoporotic fractures obviously increases with age (Honig, 2010). In addition,

candidosis (3.8%)—have been reported in allograft recipients (King et al*.*, 1994).

e. Mucosal lesions

2004). f. Oral malignancy

g. Oral infections

h. Dental anomalies

i. Bone lesions

et al*.*, 1993).

**disease**

al*.*, 2000; Squifflet and Legendre 2002).

discoloration can also occur (Wolff et al*.*, 1985).

According to a registry of residents, questionnaires were sent to all 70-year-olds among the 4,542 inhabitants of Niigata City in Japan. Participants were informed of the purpose of the survey, and the overall response rate was 81.4%. After dividing the residents into groups of males and females, 600 individuals (the screened population) were randomly selected in order to have approximately the same number of male and female participants in the study. Follow-up surveys were carried out every year in June from1998 to 2008 (11 times in 10 years), using the same methods that were used at baseline. All subjects were Japanese and did not require special care for their daily activities. Since age influences bone metabolism, renal function and periodontal disease, subjects were restricted to 70 years old at baseline (Ando et al., 2000).

#### **3.2 Osteoporosis and periodontal disease**

In addition to a strict age requirement, other study inclusion criteria included the following: blood sugar < 140 mg/dL with no history of diabetes, more than 20 teeth remaining, nonsmokers, and no history of medication use for osteoporosis. There were 184 subjects among the screened population that met all the inclusion criteria.

We utilized data on bone mineral density (BMD) of the heel, which we measured using an ultrasound bone densitometer (Fig. 1, Achilles Bone DensitometerTM, Luner Corporation,

Relationships Among Renal Function, Bone Turnover and Periodontal Disease 77

Fig. 2. Clinical attachment level and periodontal disease progression.

Fig. 3. Relationship between number of progressive sites with 3mm additional attachment

The number of subjects: stiffness 69 (n=74) and >69 (n=19) for female, 85 (n=65) and >85

A, B = Clinical attachment level B-A = Additional attachment loss

loss and stiffness by gender.

OG: Osteopenia group, NOG: No-osteopenia group.

(n=22) for male.

Fig. 1. Outline of the analysis between Osteoporosis and periodontal progression. AAL: Additional attachment loss.

USA) (Lunar Corporation, 1991). Ultrasound densitometry enables the measurement of the physical properties of bone, specifically BMD. The ultrasound measurement contains two criteria, the velocity (speed of sound, SOS) and frequency attenuation (broadband ultrasound attenuation, BUA) of a sound wave as it travels through a bone. Stiffness is a clinical index combining SOS and BUA, and is calculated by the following formula: (BUA − 50) × 0.67 + (SOS − 1380) × 0.28.

Stiffness is indicated in the bone densitometer monitoring device as the percentage of the value for a normal younger population. Osteopenia was defined as a stiffness that was 85% for males and 69% for females. Follow-up clinical surveys were done by measuring the clinical attachment level after 3 years. Clinical attachment level is the amount of space between attached periodontal tissues and a fixed point, usually the cementoenamel junction. A measurement used to assess the stability of attachment as part of a periodontal maintenance program (Fig. 2). There were 179 subjects included in the final analysis, and all of these subjects participated in both the baseline and the follow-up examinations.

We measured the number of progressive sites that had 3 mm of additional attachment loss over 3 years (Fig. 2). After dividing the subjects into an osteopenia group (OG) and a noosteopenia group (NOG), we evaluated the number of progressive sites that had 3 mm of additional attachment loss over 3 years by two-way analysis of variance (ANOVA).

The respective mean number of progressive sites for the OG and NOG were 4.75.5 and 3.33.0 in females, and 6.99.4 and 3.42.8 in males. The difference in the mean number of progressive sites between the OG and NOG was statistically significant by ANOVA after controlling for gender (Fig. 3, *p* = 0.043) (Yoshihara et al., 2004).


Fig. 1. Outline of the analysis between Osteoporosis and periodontal progression.

USA) (Lunar Corporation, 1991). Ultrasound densitometry enables the measurement of the physical properties of bone, specifically BMD. The ultrasound measurement contains two criteria, the velocity (speed of sound, SOS) and frequency attenuation (broadband ultrasound attenuation, BUA) of a sound wave as it travels through a bone. Stiffness is a clinical index combining SOS and BUA, and is calculated by the following formula: (BUA −

Stiffness is indicated in the bone densitometer monitoring device as the percentage of the value for a normal younger population. Osteopenia was defined as a stiffness that was 85% for males and 69% for females. Follow-up clinical surveys were done by measuring the clinical attachment level after 3 years. Clinical attachment level is the amount of space between attached periodontal tissues and a fixed point, usually the cementoenamel junction. A measurement used to assess the stability of attachment as part of a periodontal maintenance program (Fig. 2). There were 179 subjects included in the final analysis, and all

We measured the number of progressive sites that had 3 mm of additional attachment loss over 3 years (Fig. 2). After dividing the subjects into an osteopenia group (OG) and a noosteopenia group (NOG), we evaluated the number of progressive sites that had 3 mm of

The respective mean number of progressive sites for the OG and NOG were 4.75.5 and 3.33.0 in females, and 6.99.4 and 3.42.8 in males. The difference in the mean number of progressive sites between the OG and NOG was statistically significant by ANOVA after

of these subjects participated in both the baseline and the follow-up examinations.

additional attachment loss over 3 years by two-way analysis of variance (ANOVA).

controlling for gender (Fig. 3, *p* = 0.043) (Yoshihara et al., 2004).

AAL: Additional attachment loss.

50) × 0.67 + (SOS − 1380) × 0.28.

B-A = Additional attachment loss

The number of subjects: stiffness 69 (n=74) and >69 (n=19) for female, 85 (n=65) and >85 (n=22) for male.

OG: Osteopenia group, NOG: No-osteopenia group.

Relationships Among Renal Function, Bone Turnover and Periodontal Disease 79

To evaluate the relationship between periodontal disease and renal function markers (volume of urine per 24 hours [mL/day], creatinine clearance per 24 hours [L/day]) or bone metabolism markers (U-DPD [nM/nM\*Cr] and S-OC [ng/mL]), multiple linear regression analysis was performed. For the final model, the confounding independent variables that had *p*-values less than 0.05 according to the statistical association with the percentage of sites with 6+ mm CAL by Pearson correlation coefficients, ANOVA, or chi-square test, were selected. Results of multiple linear regression analysis between the percentage of sites with 6+ mm CAL and renal function markers after controlling for confounding factors are shown in Table 2. Creatinine clearance for 24 hours was positively associated with the percentage of sites with 6+mm CAL (sta. coef. = 0.26, *p* = 0.015). Furthermore, S-OC showed a negatively independent association with the percentage of sites with 6+vmm CAL after adjustment for

the confounding factors (sta. coef. = -0.27, *p* = 0.006, Table 3) (Yoshihara et al, 2007).

Independent variables Sta. Coef (β).\* *p* value Number of remaining teeth -0.46 <0.001 Creatinine clearance for 24 h (L/day)† 0.26 0.015 Volume of urine for 24 h (ml/day) 0.01 0.956 Smoking habit 0.08 0.500 Gender -0.17 0.121 Use of interdental brushes or dental floss -0.01 0.893 Constant 0.074

% of sites with ≥6mm attachment level

Table 2. Relationship between % of sites with ≥6mm attachment level and renal function

 % of sites with ≥6mm attachment level Independent variables Sta. Coef (β).\* *p* value Number of remaining teeth -0.47 <0.001 Serum osteocalcin (ng/ml) -0.27 0.006 Urinary deoxypyridinoline (nM/nM\*Cr) -0.04 0.688 Smoking habit -0.10 0.406 Gender 0.10 0.481 Use of interdental brushes or dental floss -0.01 0.861 Constant <0.001

Table 3. Relationship between % of sites with ≥ 6mm attachment level and bone metabolism

markers controlling for confounding factors by multiple regression analysis.

markers controlling for confounding factors by multiple regression analysis.

Dependent variable

Dependent variable

†Creatinine (g/day) in urine per 24h/creatinine (g/L) in serum.

\* Standardized coefficient.

\* Standardized coefficient.

#### **3.3 Bone metabolism and periodontal disease**

A total of 398 subjects who turned 70 in 1998 had annual dental examinations. We selected 148 of these 398 subjects (79 males and 69 females) for participation in the study because they had one or more teeth, were not taking any medicine or supplements for bone disorders (tamoxifen, anabolic steroids, bisphosphonate, or estrogen), and did not have a diagnosis of fracture based on an X-ray assessment by a physician. The subject's blood was taken in the morning of the dental examination. Urine was collected over 24 hours (07:00 to 07:00 AM the day after the dental examination). During the day that urine was collected, usual food and fluid intake were ingested. Biochemical parameters of bone turnover were measured, including urinary deoxypyridinoline (U-DPD) (nM/nM\*Cr) as a bone resorption marker, and serum osteocalcine (S-OC) (ng/mL) and serum bone alkaline phosphatase (S-BAP) (U/L) as bone formation markers. U-DPD data were corrected by the urinary creatinine concentration measured by a standard colorimetric method.

We categorized subjects by tertiles according to the percentage of sites with 6 mm clinical attachment level (6+ mm CAL). S-OC, S-BAP, and U-DPD were evaluated by analysis of covariance (ANCOVA) adjusted for smoking habit (0: none, 1: past or current). Differences in the distribution of bone turnover markers according to the percentage of sites with 6+ mm CAL per person are shown in Table 1. S-OC was significantly lower in the third tertile than in the first and second tertiles after adjusting for smoking habit (males: *p* = 0.007, females: *p* = 0.042, ANCOVA) (Yoshihara et al., 2009).


\* ANOCOVA adjusted for smoking habits.

Table 1. Relationship between % of sites with ≥ 6mm attachment level and bone metabolism markers controlling for confounding factors by multiple regression analysis.

#### **3.4 Renal function and periodontal disease**

We randomly selected 145 subjects among 398 healthy elderly subjects. All subjects were aged 77 years at the time of the renal function study in 2005. We evaluated the relationship between bone turnover markers and periodontal disease, taking renal function into consideration. Correlations among renal function and bone metabolism markers for periodontal disease, including the number of remaining teeth and smoking habit, were evaluated using multiple regression analysis.

To evaluate the relationship between periodontal disease and renal function markers (volume of urine per 24 hours [mL/day], creatinine clearance per 24 hours [L/day]) or bone metabolism markers (U-DPD [nM/nM\*Cr] and S-OC [ng/mL]), multiple linear regression analysis was performed. For the final model, the confounding independent variables that had *p*-values less than 0.05 according to the statistical association with the percentage of sites with 6+ mm CAL by Pearson correlation coefficients, ANOVA, or chi-square test, were selected. Results of multiple linear regression analysis between the percentage of sites with 6+ mm CAL and renal function markers after controlling for confounding factors are shown in Table 2. Creatinine clearance for 24 hours was positively associated with the percentage of sites with 6+mm CAL (sta. coef. = 0.26, *p* = 0.015). Furthermore, S-OC showed a negatively independent association with the percentage of sites with 6+vmm CAL after adjustment for the confounding factors (sta. coef. = -0.27, *p* = 0.006, Table 3) (Yoshihara et al, 2007).


†Creatinine (g/day) in urine per 24h/creatinine (g/L) in serum.

\* Standardized coefficient.

78 Chronic Kidney Disease

A total of 398 subjects who turned 70 in 1998 had annual dental examinations. We selected 148 of these 398 subjects (79 males and 69 females) for participation in the study because they had one or more teeth, were not taking any medicine or supplements for bone disorders (tamoxifen, anabolic steroids, bisphosphonate, or estrogen), and did not have a diagnosis of fracture based on an X-ray assessment by a physician. The subject's blood was taken in the morning of the dental examination. Urine was collected over 24 hours (07:00 to 07:00 AM the day after the dental examination). During the day that urine was collected, usual food and fluid intake were ingested. Biochemical parameters of bone turnover were measured, including urinary deoxypyridinoline (U-DPD) (nM/nM\*Cr) as a bone resorption marker, and serum osteocalcine (S-OC) (ng/mL) and serum bone alkaline phosphatase (S-BAP) (U/L) as bone formation markers. U-DPD data were corrected by the urinary

We categorized subjects by tertiles according to the percentage of sites with 6 mm clinical attachment level (6+ mm CAL). S-OC, S-BAP, and U-DPD were evaluated by analysis of covariance (ANCOVA) adjusted for smoking habit (0: none, 1: past or current). Differences in the distribution of bone turnover markers according to the percentage of sites with 6+ mm CAL per person are shown in Table 1. S-OC was significantly lower in the third tertile than in the first and second tertiles after adjusting for smoking habit (males: *p* = 0.007,

1st 2nd 3rd *p* value\* 1st 2nd 3rd *p* value\*

(ng/ml) 8.5 ± 4.5 6.8 ± 2.7 5.7 ± 1.8 0.007 9.9 ±2.8 9.3 ± 2.4 9.1 ± 3.5 0.042

phosphatase (U/L) 22.2 ± 5.9 23.3 ± 7.4 21.1 ± 6.2 0.212 29.3 ± 10.8 28.9 ± 8.1 27.4 ± 11.2 0.752

Table 1. Relationship between % of sites with ≥ 6mm attachment level and bone metabolism

We randomly selected 145 subjects among 398 healthy elderly subjects. All subjects were aged 77 years at the time of the renal function study in 2005. We evaluated the relationship between bone turnover markers and periodontal disease, taking renal function into consideration. Correlations among renal function and bone metabolism markers for periodontal disease, including the number of remaining teeth and smoking habit, were

markers controlling for confounding factors by multiple regression analysis.

4.8 ± 1.0 4.4 ± 1.2 4.0 ± 1.0 0.055 6.6 ± 1.4 6.8 ± 1.4 6.3 ± 1.7 0.664

creatinine concentration measured by a standard colorimetric method.

females: *p* = 0.042, ANCOVA) (Yoshihara et al., 2009).

% of sites with 6mm attachment level

Serum osteocalcin

Serum bone alkaline

deoxypyridinoline (nM/nM\*Cr)

\* ANOCOVA adjusted for smoking habits.

**3.4 Renal function and periodontal disease** 

evaluated using multiple regression analysis.

Urinary

Males Females

**3.3 Bone metabolism and periodontal disease** 

Table 2. Relationship between % of sites with ≥6mm attachment level and renal function markers controlling for confounding factors by multiple regression analysis.


\* Standardized coefficient.

Table 3. Relationship between % of sites with ≥ 6mm attachment level and bone metabolism markers controlling for confounding factors by multiple regression analysis.

Relationships Among Renal Function, Bone Turnover and Periodontal Disease 81

Fig. 4. The mechanism between low renal function and periodontal disease.

c. CRF has an important effect on vitamin D metabolism (Yoshihara et al., 2007). Since vitamin D is metabolized in the liver and kidney, the presence of CRF will automatically disturb vitamin D metabolism. Vitamin D is metabolized by kidney to its active metabolite, 1,25-dihydroxyvitamin D3. This substance subsequently interacts with vitamin D nuclear receptor in the intestine, bone and kidney. The functions of this substance are to regulate bone metabolism, immune response and also cell proliferation and differentiation. Regarding bone metabolism, vitamin D controls the availability of calcium phosphate by regulating the excretions of hormones such as the parathyroid hormone (PTH) (Souza et al., 2007). CRF may disrupt the regulation of PTH which may leads to hyperparathyroidism condition and increased rate of bone disease (Yoshihara et al., 2007). Vitamin D also contributes in the synthesis of bone matrix proteins such as type-I collagen, alkaline phosphatase, osteocalcin and osteopontin (Souza et al., 2007). Osteocalcin may exist in the circulating blood and undergo local accumulation in some parts of the body. Osteocalcin has been postulated to have a role in both bone resorption and mineralization and is currently considered the most specific marker of osteoblast function. The serum level of this protein is considered to be a marker of bone formation. Serum osteocalcin is presently considered a valid marker of bone turnover

The results showed that the subjects in the OG had a higher number of progressive sites for additional attachment loss than the subjects in the NOG. This three-year longitudinal study clearly demonstrated that BMD is a risk factor for periodontal disease progression in an elderly population. In addition, according to our findings on linkage with BMD, there are some systemic factors that contribute to both loss of bone mass and periodontal disease progression (Kshirsagar, 2005). Systemic factors of bone remodeling may also modify the local tissue response to periodontal disease. The BMD of the mandible is affected by the mineral status of the skeleton and also by diseases that cause generalized bone loss (Davidovich, 2005). The mouth and face are highly accessible parts of the body, and reflect changes that occur internally. For the clinician, the mouth and face provide physical signs and symptoms of local and generalized disease. During routine oral examinations, periodontal disease including maxillary/mandibular general bone loss may be diagnostic of early osteoporotic changes in the skeleton. Some systemic factors of bone remodeling also modify the local tissue response to periodontal disease.

Osteoporosis and low renal function contribute to loss of bone mass. We were able to identify a weak but clear relationship between CAL and S-OC. There was a significant association between CAL and 24-hour creatinine clearance, which is a renal function marker**.** These findings suggest that S-OC is a valid marker of bone turnover when evaluating periodontal disease. It has been assumed that S-OC is associated with not only bone turnover but also low renal function. Periodontal conditions, including bone metabolism, may be affected by low renal function. The systemic bone metabolism, which might be affected by low renal function, is associated with periodontal disease.
