Oral Microbiota, Periodontal Status, and Osteoporosis in Postmenopausal Females

Background: Alterations in hormone levels during menopause decrease bone density and may worsen oral health, favoring the growth of periodontal pathogens, whose detection could improve the diagnosis of periodontitis. The aim of this study is to detect and quantify the main periodontal pathogens in the oral microbiota of postmenopausal females and to explore the relationship between clinical and periodontal parameters. Methods: This was an observational cross-sectional study of 76 postmenopausal females. Dental examinations and sampling for microbiologic evaluation were performed, and a history of osteoporosis/osteopenia was collected. Real-time polymerase chain reaction was used for detecting and quantifying Fusobacterium nucleatum (Fn), Aggregatibacter actinomycetemcomitans (Aa), Porphyromonas gingivalis (Pg), Campylobacter rectus (Cr), and Tannerella forsythia (Tf). The results obtained were subjected to statistical analyses. Statistical significance was defined as P <0.05. Results: Periodontitis was detected in 77.1% of females with osteoporosis/osteopenia (P >0.05). A significant correlation was found between osteoporosis and missing teeth. T. forsythia and C. rectus were detected in 100% of the samples, Fn and Pg in 98.7%, and Aa in 73.7%. Conclusions: Osteoporosis did not influence the prevalence of periodontitis among postmenopausal females. The presence of periodontopathogenic bacteria was not sufficient to confirm disease. A preventive maintenance program for postmenopausal females, particularly osteoporotic females, who are at greater risk of tooth loss, could minimize the potential effects of bone loss on periodontal tissues. J Periodontol 2016;87:124-133.


M
enopause is the permanent cessation of menstrual periods, whether naturally or induced by surgery, chemotherapy, or radiation.It is accompanied by hormonal changes, most prominently decreases in estrogen and androgen levels, which produce a variety of signs and symptoms: 1) vasomotor symptoms; 2) decreased bone density; 3) hot flashes; 4) profuse perspiration; 5) atrophic vaginitis; and 6) oral cavity alterations. 1 The oral symptoms during menopause include dry mouth (xerostomia), burning mouth, and changes in taste perception and are a common complaint of postmenopausal females. 2 If xerostomia is left untreated, it may change the oral pH and increase the risk of developing oral diseases.These symptoms have been attributed to reduced salivary flow, and oral pathologies seem to be affected by changes in salivary and gingival crevicular fluid (GCF) composition. 3Saliva contributes to maintenance of the oral pH by neutralizing acids from food and beverages, as well as from bacterial activity, thereby reducing the risk of periodontal disease. 3After menopause, reduced salivary flow rate and low pH may make females more prone to oral health problems. 4Decreased estrogen levels, in addition to worsening periodontal conditions, increase the risk of osteoporosis in postmenopausal females. 3Osteoporosis is a systemic skeletal disease characterized by a generalized and progressive loss of bone mineral density (BMD) and an increased propensity to fracture. 5,6Its prevalence increases with age, and it affects all bones of the body, including the jaw bone, resulting in cortical thinning and a reduction of cancellous bone. 6or females, in addition to age, the menopausal transition itself causes a period of rapid bone loss (12% to 20% in the first 5 years after menopause) such that the risk of fracture in postmenopausal females is higher than in males of the same age. 7he changes in bone composition also have implications for oral health, 8 including the risk of enhanced progression of periodontitis. 5,6Thus, bone loss is a feature shared by periodontal disease and osteoporosis.The relationship between the skeletal loss of BMD and the increased periodontal bone loss has been ascribed to several factors. 5Greater periodontal bone loss may occur simply because the bone surrounding the teeth is less dense and therefore less resistant to resorption. 5 In addition, the loss of alveolar BMD may leave bone more susceptible to periodontal bacteria, increasing the risk for periodontitis and tooth loss. 9steoporotic females show greater clinical attachment loss and gingival recession than females with a normal BMD, which suggests a greater severity of periodontitis. 10 The early diagnosis of reduced BMD, before the establishment of a significant negative influence on the periodontal tissues, has been recommended. 9he pathogenic microbial population involved in periodontitis is highly complex.However, the literature contains reports of associations among periodontal disease and certain microorganisms, such as Aggregatibacter actinomycetemcomitans (Aa), Prevotella intermedia (Pi), Porphyromonas gingivalis (Pg), Campylobacter rectus (Cr), Treponema denticola, Fusobacterium nucleatum (Fn), and Tannerella forsythia (Tf). 11,12A quantitative analysis of these periodontal pathogens could improve both the diagnosis and evaluation of periodontitis and the understanding of its relationship to osteoporosis.
The aim of this study is to detect and quantify the main periodontal pathogens in the oral microbiota of postmenopausal females to explore the relation between the microbiota and clinical and periodontal parameters.The measurement of pH was done as a general parameter because it is related strongly with consumption of sugar and acidic beverages, xerostomia, and oral microbiota. 3

Study Population
This is an observational cross-sectional study of 76 postmenopausal females who were treated at the Dental Hospital of the University of Barcelona, Barcelona, Spain.Dental examinations and sampling for microbiologic evaluation were performed between May 2009 and September 2012.
The study population was selected from 694 females who participated in a questionnaire related to fracture risk factors in the Rheumatology Service of the University Hospital of Bellvitge, Hospitalet, Barcelona, Spain, in 1999.Only 137 of the 427 females who could be contacted agreed to participate.Subgingival samples were collected from 76 of these females, who were finally included in the study (aged 40 to 71 years; median age: 63 years).None of them were under specific estrogen replacement treatment.
The study was approved by the Ethical Committee of the Faculty of Dentistry, University of Barcelona.The study was conducted in accordance with the Declaration of Helsinki of 1975 (revised in 2013).All of the patients were informed about the objectives and methodology and provided written informed consent to participate in the study.

Oral Examination
All of the participants received a complete oral examination.The following variables were assessed: 1) number of missing teeth, not including third molars (28 maximum); 2) probing depth (PD), assessed with a World Health Organization (WHO) probe in all teeth; 3) oral hygiene status, determined according to the criteria of the index of Greene and Vermillion; 13 4) the carious, absent, and obturated teeth (CAOD) index, which was obtained by summing carious, missing, and obturated teeth and dividing by the total number of teeth; 5) the presence of mucosal injury; 6) periodontal treatment needs, examined by using the Community Periodontal Index of Treatment Needs recommended by the WHO; 7) oral pH, measured by using pH indicator strips ¶ that were placed on the floor of the mouth; and 8) xerostomia, determined by using a specific questionnaire and performing the test of the mirror (sticking of a dental mirror to the buccal mucosa was taken as a positive result for xerostomia).

Diagnosis of Osteoporosis/Osteopenia
Osteopenia was assessed from densitometry reports.The methods and criteria used in the diagnosis of osteoporosis were those established by the WHO. 14ccording to these criteria, osteoporosis is defined as BMD ‡2.5 standard deviations (SDs) below the average value for young healthy females (T score £2.5 SDs), and osteopenia is defined as 1 to 2.5 SDs (T score between -1.0 and -2.5 SDs).Association between osteopenia and oral bone loss and between osteopenia and oral state were accomplished as recommended in the literature. 15,16iagnosis of Periodontitis Females were considered to have periodontitis if they met the criteria established by the American Academy of Periodontology, which state that the patient must have at least one site with PD and clinical attachment level ‡4 mm. 17ecimen Collection Subgingival samples were taken after supragingival plaque removal.Two paper points (size 30) # were inserted in the deepest periodontal pocket for 20 seconds and placed in labeled sterile test tubes containing 500 mL bidistilled water.The vials were immediately frozen at -80°C.Despite the limitation derived from the sampling of only one pocket, the samples were harvested from the deepest one, because deeper pockets offer a better environment for anaerobes and are useful as a general parameter.

DNA Extraction
The tubes containing the paper points were thawed, vortexed for 2 to 3 minutes, and then centrifuged at 14,000 rpm for 10 minutes.The pellet was processed with a commercial DNA extraction kit** following the protocol of the manufacturer for Gram-negative bacteria.DNA purity and concentrations were determined spectrophotometrically using a spectrophotometer.† † DNA from the controls (reference bacterial strains) was obtained using the same method.Purified genomic DNA of Pg, Aa, Tf, Cr, and Fn was obtained from 1 mL viable bacteria containing 10 8 , 10 9 , 10 8 , 10 6 , and 10 6 colony forming units (CFU)/mL, respectively.Tenfold serial dilutions of each DNA were prepared in sterile water.‡ ‡ All DNA samples were frozen at -20°C until used.

Quantitative Real-Time Polymerase Chain Reaction
Primers and probes.The primers § § and probes ii sequences used to target the 16S ribosomal ribonucleic acid (rRNA) gene are listed in Table 1. 18,19The oligonucleotide probes were labeled at the 59 end with fluorescent 6-carboxyfluorescein and at the 39 end with 6-carboxytetramethylrhodamine.
Quantitative real-time polymerase chain reaction conditions.Polymerase chain reaction (PCR) amplification was performed in a total reaction mixture volume of 10 mL, containing 8 mL of 2• master mixture ¶ ¶ and 2 mL DNA from the samples.Each DNA sample was analyzed in duplicate.The no template control (NTC) consisted of 8 mL master mixture without DNA.The samples were placed in white 384-well plates ## and subjected to an initial amplification cycle of 95°C for 10 minutes, followed by 40 cycles at 95°C for 15 seconds and 60°C for 1 minute in a thermocycler.***To avoid carryover DNA, the laboratory was separated into pre-PCR and post-PCR areas, and strict barrier measures were used.However, none of these methods prevented the amplification of unspecific bacteria in some of the NTCs.
Data analyses.Quantification cycle (Cq) values were determined using a software package.† † † Quantitation of the DNA samples was based on accurate standard curves for each bacterium in each run, using appropriate external standards of known concentration.Standard curves were constructed by plotting the Cq values generated from the quantitative PCR (qPCR) against Pg, Tf, Cr, Fn, and Aa (see supplementary Figures 1 through 5 in online Journal of Periodontology) and total cell concentrations (log CFU/mL).The correlation between Cq values and CFU/mL was automatically generated.
All assays were developed with a linear quantitative detection range established by a slope of 3.2 to 3.7 cycles/log decade, r 2 >0.994 and an efficiency range of 1.8 to 2.0.The limit of detection was calculated using the Cq value from the last point on the standard curve that differed by 3 units from the lowest Cq value of the NTC.Values below this detection range may be detectable, but they are not quantifiable.Samples were considered positive when the Cq values were above the detection limit.

Statistical Analyses
A descriptive statistical analysis was performed based on the mean or median according to normality criteria (Kolmogorov-Smirnov test) and the SD or interquartile range (IQR), respectively.A bivariate analysis was performed using the x 2 test for the comparison of qualitative variables.Student t test, analysis of variance, the Mann-Whitney U test, or the Kruskal-Wallis test for quantitative variables was used according to normality criteria and number of categories studied.Pearson or Spearman correlation tests were used to compare quantitative variables when appropriate.A multivariate analysis using binary logistic regression was conducted to determine the association between osteoporosis and periodontitis.The statistical significance of the two-way interactions between BMD and other variables in the main effects model was assessed using forward stepwise regression to test whether other variables modified the effect of BMD on periodontitis.To select the most accurate model, the area under the curve of each model was analyzed according to the receiving operating characteristics curves.A statistical package ‡ ‡ ‡ was used for all statistical analyses.Statistical significance was defined as P <0.05.

RESULTS
The general characteristics and relevant variables regarding the oral health status of the 76 females who participated in the study are listed in Table 2.For some females, the data were incomplete because they could not attend all sessions of data collection.All participants were kept in the analysis because the percentage of missing data was <10% of the total for all variables.
The participants had a median age of 63.0 years (IQR = 2).Periodontitis was detected in 47.9% of the patients, and 41.1% required periodontal treatment.Most (78.6%) had been diagnosed previously with osteopenia or osteoporosis.
Table 3 provides a cross-tabulation of periodontal disease versus osteoporosis, age, number of missing teeth, oral pH, CAOD index, need for treatment, oral hygiene, xerostomia, and mucosal injury.Periodontitis was detected in 22.9% of the females with no previous diagnosis of osteopenia/osteoporosis, in 57.1% of the females with osteopenia, and in 20% of those with osteoporosis, although there was no association between osteoporosis/osteopenia and periodontitis (P = 0.52).There was a significant correlation between periodontitis and oral hygiene (P = 0.003) and between periodontitis and the need for treatment (P = 0.001).The associations between periodontitis and missing teeth and between periodontitis and xerostomia were not significant (P = 0.12 and P = 0.10, respectively).
An analysis of the correlation among osteoporosis and the same clinical variables (data not shown) showed that, among the two groups (normal/osteopenia and osteoporosis), only the relation to the number of missing teeth was significant (P <0.05).The most prevalent bacteria were Tf and Cr, which were detected in 100% of the samples, followed by Fn and Pg (98.7% in both cases) and Aa (73.7%).It should be highlighted that, in all cases but Pg, it was feasible to quantify bacteria when present.Pg was quantifiable in only 84.2% of the samples, because in 14.5% the value was below the minimum needed to quantify.Quantification of periodontal bacteria according to the presence or absence of periodontitis is reported in Table 4.
The box plot in Figure 1A shows the distribution of the five periodontal bacteria according to the presence or absence of periodontitis; the differences were statistically non-significant (P >0.05).The box plot in Figure 1B shows the distribution of bacteria in osteo-porotic, osteopenic, and normal patients.Again, the differences between the groups were not significant (P >0.05).
Scatter plots were used to analyze the relationship between the periodontal pathogens Fn, Tf, Cr, Pg, and Aa, and non-parametric Spearman correlation coefficients were calculated (Table 5).Statistically significant positive relationships were found between log Cr and log Fn (r = 0.815, P <0.001) and log Tf and log Fn (r = 0.510, P <0.001).Other significant relationships among bacteria included a moderately positive relationship between log Pg and log Tf (r = 0.410, P <0.001) and slightly positive relationship between log Tf and log Cr (r = 0.381, P = 0.001), log Pg and log Cr (r = 0.328, P = 0.004), log Pg and log Fn (r = 0.268, P = 0.02), and log Aa and log Tf (r = 0.273, P = 0.04).
Spearman correlation was used to compare periodontal pathogens and oral parameters.A significant highly positive relationship was found between missing teeth and CAOD index (r = 0.770, P <0.001) and a slightly positive relation between log Pg and pH (r = 0.285, P = 0.02).
Analysis of the associations between clinical variables and pH values (data not shown) failed to demonstrate a significant correlation between oral hygiene and pH (P = 0.50).Only the differences among basic, neutral, and acidic pH groups and the log CFU/mL of Aa were significant (P = 0.01).For Pg, the P value was 0.053.

DISCUSSION
According to the present findings, the presence of osteoporosis/osteopenia does not influence the prevalence of periodontitis among postmenopausal females.The criteria for the diagnosis of periodontitis were clinical examination parameters.In a systematic review by Martı ´nez-Maestre et al., 20 five of nine studies based on the same criteria concluded that the correlations between osteoporosis and periodontitis were negative.In contrast, in most of the studies based on maxillary radiologic criteria, a positive association was determined.Martı ´nez-Maestre et al. 20 highlighted the need for well-controlled studies to better determine the complexity of this relationship.
The results of two recent studies performed on samples from postmenopausal females also reached conflicting conclusions, thus adding to the current controversy and emphasizing the need for greater uniformity in the methodologies to study the relationship between osteoporosis and periodontitis. 21,22ethodologic limitations include small sample sizes, non-comparable study populations, and the diverse criteria used to define osteoporosis and periodontitis.
Although there was no correlation between both diseases, when the patients were evaluated in groups of normal/osteopenia and osteoporosis, significant differences were found with respect to the number of missing teeth.The associations between dental status and skeletal bone density in healthy postmenopausal females have been investigated previously.Some findings support the hypothesis that systemic bone loss contributes to tooth loss. 23ooth loss is considered the ultimate outcome of periodontal disease. 24The chronic inflammation triggered by the bacteria that cause periodontal disease underlies both breakdown of the jaw bone and tooth loss.Tezal et al. 25 reported that periodontal disease is a strong and independent predictor of tooth loss in postmenopausal females.However, in the patients in the present study, a relationship between periodontitis and missing teeth could not be established.7][28] For example, the systemic statins frequently used in lipid-lowering therapies have an anti-inflammatory effect on periodontal disease and protect against tooth loss. 29Higher rates of tooth loss have also been associated with low socioeconomic status, a low level of education, and low income and with age and smoking habits. 30However, this information was not collected from the participants in the present study.The results showed that, in the present patients, periodontitis was correlated with the level of oral hygiene.Therefore, the potential effects of osteoporotic factors on the periodontal tissues could be minimized by combining good oral hygiene with regular check-ups.The association between periodontitis and xerostomia was not significant, but a positive trend was evident because five of the six females with dry-mouth symptoms were diagnosed with periodontitis.More patients with xerostomia are required to better analyze this relationship.] In this study, qPCR assays are used successfully to determine the abundance of periodontal pathogens known to be associated with the progression and severity of periodontal disease in paper-point samples collected from postmenopausal females.The samples were harvested only from the deepest pocket, because deeper pockets offer a better environment for anaerobes and, consequently, their more extensive colonization and are useful as a general parameter.The prevalences of four of the five periodontal bacteria analyzed (Pg, Tf, Cr, and Fn) were very high (98.7% to 100%), and Aa was detected in 73.7% of the patients.In a similar study, Brennan et al. 33 determined the prevalence of infection with periodontal bacterial species in a sample of postmenopausal females.They found Tf in 37.9% of the samples and Cr, Pg, and Fn in 14% to 17%.Disparities between study methods may account for the differences in their sensitivities with respect to the detection and quantification of periodontal pathogens.Brennan et al. 33 used indirect immunofluorescence microscopy, which has important drawbacks compared with quantitative PCR, including a limited capability for accurate quantification and higher limits of detection. 34acuch and Tanner 35 identified Cr in 90% of their patients with newly diagnosed and established periodontitis, in 20% of those with gingivitis, and in 10% of those considered healthy.However, the samples in that study originated from a different source population than in the present study and relied on bacterial culturing methods, which have a lower sensitivity of detection. 35he sensitivity and specificity of qPCR are higher because this method overcomes most of the limitations of the other techniques. 34,36,37Therefore, the use of qPCR is one of the strengths of the present work.Riep et al. 36 examined the prevalence of 10 periodontal bacterial species in three groups of patients, those with CP or generalized aggressive periodontitis and those in a periodontitis-resistant control group.PCR of 16S rRNA gene fragments and subsequent dot-blot hybridization were used as the detection method.The prevalence of Tf, Pg, and Pi was high in all groups (60% to 95% of positive patients), without significant differences among them, although these species are considered to be highly associated with advanced periodontitis. 31The prevalence of some of the target species in that study are similar to those detected in this study.
In examining the potential correlation between bacterial species, significant associations were found between Fn and Cr; this will fit with the orange complex of Socransky et al., 37 Fn and Tf, and Pg and Tf (red complex), among others.The simultaneous presence of these bacteria suggests their interdependence, which could further promote the progression of periodontal disease.
The lower presence of Aa detected in this study is correlated with an acidic oral environment.These results are in agreement with the observed optimal pH range for growth of this bacterium. 38The findings also suggest a greater abundance of Pg under acidic conditions (pH below 7), although optimum pH for growth is close to neutrality. 39Maybe the association could be attributable to low salivary flow more than to pH.This information could help clinicians choose the best approach for treating periodontal infections caused by these microorganisms.
The bacterial species investigated in this study do not show significant quantitative differences among Table 5. healthy, osteopenic, and osteoporotic patients.The correlations between females with periodontitis and healthy patients were also negative.Therefore, the presence of subgingival periodontopathogenic bacteria in almost all of the present patients demonstrates that the appearance of these species is not sufficient to confirm the presence of periodontal disease.In fact, the mere presence of oral pathogens is recognized as being necessary but not sufficient for disease occurrence.1][42] Still, disease will not necessarily develop as long as the levels of the pathogen do not exceed the threshold for the host. 38In the present study, some of the healthy patients may have had early-stage, albeit, by the criteria of this study, non-detectable disease, which may have blunted any significant differences between patients with periodontitis and those who were apparently healthy.

CONCLUSIONS
In the present study, osteoporosis did not influence the prevalence of periodontitis among postmenopausal females.The presence of periodontopathogenic bacteria was not sufficient to confirm disease, although the influence of PD on the prevalence of the pathogens analyzed was not examined.Probably, the detection of these bacterial species may be more closely related to the depth of the sampled pocket than to the diagnosis of periodontitis, because deeper pockets offer a better environment for anaerobes and, consequently, their more extensive colonization. 36,43A preventive maintenance program for postmenopausal females, particularly osteoporotic females, who are at greater risk of tooth loss, could minimize the potential effects of bone loss on periodontal tissues.Additional studies are needed to examine the influence of confounding factors for periodontitis and osteoporosis in postmenopausal females.

Figure 1 .A)
Figure 1.A) Box plot of normal patients and the periodontitis group and quantification of periodontal bacteria.B) Box plot of normal, osteopenic, or osteoporotic patients and quantification of periodontal bacteria.

Table 1 .
Primers and Probes Used for Quantification of Genomic DNA From the Target Bacteria Primers and probes targeted the 16S rRNA gene.‡ ‡ ‡ SPSS v.17.0 for Windows, IBM, Chicago, IL.

Table 2 .
Characteristics of the Patients (n 5 76)

Table 3 .
Correlation Among Periodontitis and Clinical and Oral Variables

Table 4 .
Quantification of Subgingival Pg, Tf, Cr, Fn, and Aa in the Total Study Sample (N 5 76)