Obesity as a modifying factor of periodontal therapy outcomes: local and systemic adipocytokines and oxidative stress markers

Adipocytokines and oxidative stress (OS) are involved in the pathogenesis of both obesity and periodontitis. The aim of this study was to evaluate periodontal therapy outcomes in terms of serum and gingival crevicular fluid (GCF) levels of adipocytokines and OS markers in obese patients with periodontitis, in order to have an insight into the association between obesity and periodontitis. A total of 39 patients (20 obese, 19 non-obese) with periodontitis were included in this study. Clinical periodontal parameters were assessed; serum and GCF levels of adipocytokines and OS markers were evaluated by ELISA at baseline and 3 months after non-surgical periodontal therapy. Significant improvements in clinical periodontal parameters were observed in both groups at 3 months (p < 0.01). While serum levels of TNF-α, leptin, and total oxidant status (TOS) in the obese group were higher at baseline (p < 0.01), leptin levels remained higher at 3 months despite a significant decrease (p < 0.01). Although NSPT improved GCF levels of total antioxidant status (TAS) and TOS in both groups, they were significantly different between the groups after therapy (p < 0.05). It seems that leptin, TNF-α, and TOS contribute to systemic inflammatory and oxidative state in patients with obesity. Despite improvements in clinical periodontal parameters, obesity might be a modulating factor in the development and progression of periodontal disease in terms of some adipocytokines and OS markers. Since the global burden of both obesity and periodontitis is continuously increasing, the management of these inflammatory diseases has become more important. The current study contributes to our understanding of the role of OS and adipocytokines on the relationship between obesity and periodontitis by response to periodontal treatment.


Introduction
Obesity is a chronic, inflammatory, multi-factorial disease defined by the World Health Organization as "abnormal or excessive fat accumulation in the body that may impair health" [1]. Obesity, induced by a sedentary lifestyle and a high-fat diet, has become a major public health problem and a worldwide epidemic, affecting 52% of the world's population. The prevalence of obesity has increased almost threefold between 1975 and 2016, with over 1.9 billion overweight and 650 million obese adults worldwide [1]. Although adipose tissue was considered a passive energy store until recently, it has been determined to be a metabolically effective organ secreting many bioactive molecules known as adipocytokines that regulate immunity and inflammation [2]. Furthermore, obesity is a state of oxidative stress 1 3 (OS) characterized by the overproduction of reactive oxygen species (ROS).
Periodontitis is a chronic inflammatory, multi-factorial disease of tooth-supporting structures, which can be modified by systemic, environmental, and genetic factors [3]. It has been reported that severe periodontitis is the sixth most common chronic disease worldwide, affecting approximately 743 million people [4]. After being initiated by dysbiotic dental plaque, periodontal destruction occurs as a result of unusual host response exhibiting excessive inflammation with ROS release. In case of an inflammatory environment, the dysbiotic plaque further develops and stimulates inflammatory responses [5]. Studies have suggested an association between periodontitis and systemic disorders, including obesity. Although the possible relationship between obesity and periodontitis has been emphasized by many studies, its mechanism has not been exactly clarified [6].
The potential biological mechanisms shared by both obesity and periodontal disease are hyperinflammation, altered immune response, dysbiosis, genetic polymorphisms, and even stress [7]. While the imbalance between increased proinflammatory adipocytokines and decreased anti-inflammatory mechanisms in obesity leads to chronic low-grade inflammation, adipocytokines induce inflammatory processes and OS disorders, creating a similar pathophysiology between obesity and periodontitis [8]. It has been reported that insulin resistance as a result of chronic inflammatory status and OS may play a role in the relationship between obesity and periodontitis. In addition, increased host immune and inflammatory responses due to systemic low-grade inflammation can be considered as a risk for the development of periodontitis [9].
The studies reveal that OS, resulting from an imbalance between ROS and antioxidant capacity, is associated with obesity as well as periodontitis [10]. Although there is much data about the effects of adipocytokines on the relationship between obesity and periodontitis and on response to periodontal treatment, studies examining the role of OS are scarce as if there are not any [11]. Furthermore, no study has evaluated local and circulating levels of both adipocytokines and OS markers in response to non-surgical periodontal therapy (NSPT). Thus, the aim of this study was to investigate the effects of periodontal therapy on clinical response, gingival crevicular fluid (GCF), and serum levels of adipocytokines and OS markers in obese and non-obese patients with periodontitis.

Materials and methods
Twenty-three obese and 20 non-obese patients diagnosed with periodontitis (stage II/III) [3] who were referred to the Department of Periodontology, Faculty of Dentistry, Inonu University, were included in the study. The study was approved by the Ethical Committee of Malatya Clinical Researches, Inonu University (Protocol 2014/40) and was conducted according to the principles of the Helsinki Declaration revised in 2013. All patients were informed about the study and gave their written consent to participate.

Patient selection
Diagnosis of obesity was made considering these criteria: body mass index (BMI; calculated as weight divided by the square of height (kg/m 2 )) as a sign of total adiposity and waist circumference (WC) indicating abdominal adiposity [12]. Patients with BMI ≥ 30 kg/m 2 and WC > 88 cm for females and > 102 cm for males were defined as obese. They were advised not to change their diets throughout the study. At the end of the study, anthropometric measurements were reassessed to determine their status of being obese.
Exclusion criteria were systemic problems that might affect the course of periodontitis (e.g., DM, metabolic syndrome, osteoporosis, and immunological disorders), smoking, periodontal treatment within the previous 12 months, history of anti-inflammatory, anti-microbial, anti-oxidant, or immunosuppressive therapy within the previous 6 months, being during pregnancy or lactation.

Study design and sample size calculation
This study was designed as a single-center, controlled, fullmouth, 3-month clinical trial and is presented in Fig. 1. According to the study protocol, individuals selected with respect to the above criteria were grouped after radiographic and clinical examinations.
• Group C; non-obese patients with periodontitis, (8 women, 12 men; total of 20 individuals with a mean age of 44.80 ± 7.52) • Group OB; obese patients with periodontitis, (14 women, 9 men; total of 23 individuals with a mean age of 47.52 ± 6.36) Sample size was calculated by power analysis regarding a difference of 0.42 pg/µl for TNF-α levels between groups and a standard deviation of 0.28 pg/µl, based on a previous study [13]. It was determined that at least 16 volunteers were required for each group to provide a statistical power of 90% and error α as 0.01. Regarding the possibility of dropouts, 20 and 23 patients were included in groups C and OB respectively. Since 3 obese and 1 non-obese patients did not keep on, the study was completed with 20 obese and 19 non-obese patients.

Clinical procedures
Oral hygiene instructions were given to all patients one week before treatment and it was checked whether they were successful in maintaining plaque control during the study. Full-mouth clinical measurements (plaque index (PI), gingival index (GI), bleeding on probing (BOP), probing depth (PD), and clinical attachment level (CAL)) were performed at baseline and 3 months after periodontal treatment by one trained, calibrated examiner (VET). PD and CAL were assessed at six sites (mesio/disto-buccal, mid-buccal, mesio/disto-lingual, and mid-lingual) per tooth, except third molars, using Williams periodontal probe (Hu-Friedy, Chicago, IL, USA). For examiner calibration, the examiner repeated PD measurements of 210 sites randomly in five patients 1 week apart and data were evaluated by Student t test. Differences between the measurements were not statistically significant (p > 0.05). Additionally, panoramic and periapical radiographs were taken to verify the diagnosis of periodontitis.
One week after clinical measurements, all patients received non-surgical periodontal therapy (NSPT) using an ultrasonic device (Electro Medical Systems SA, Nyon, Switzerland) and Gracey curettes (Hu-Friedy, Chicago, IL, USA) under local anesthesia. Periodontal therapy was performed within 24 h during two sequential visits. No additional treatment, such as local or systemic antimicrobials, was applied.

Serum and gingival crevicular fluid sampling
Fasting venous blood and GCF samples were collected from all individuals at baseline and 3 months. Blood samples were rested for 30 min at room temperature, centrifuged (at 4 ºC, 3500 rpm / min for 7 min) and separated serum samples were stored at − 80℃ until analyzed.
Following blood sampling, GCF samples were collected from four discontiguous, deepest pocket regions between 5 and 7 mm in depth in the upper jaw of each patient. The teeth to be sampled were isolated, dried and the supragingival plaque was removed with sterile curettes gently. Paper strips (Periopaper®, Proflow Inc, NY, USA) were advanced 1 mm in the pockets and kept for 30 s [14]. Blood/ saliva-contaminated strips were excluded. GCF volume was measured using Periotron 8000® (Oraflow Inc., Plainview, NY, USA), calibrated previously and Periotron units were converted to the actual volume (μl) according to the standard curve. Paper strips obtained from four regions of each patient were placed in Eppendorf tubes containing 200 µl Fig. 1 Flowchart of the study phosphate buffer, vortexed for 30 s at room temperature, and centrifuged at 3000 g for 5 min to remove cellular elements and plaque. GCF samples were stored at − 80 ºC until ELISA (Enzyme-Linked Immunosorbent Assay) analyses were performed.
Detection of adipocytokines and OS markers was done by ELISA using commercial kits for tumor necrosis factor-α (TNF-α), 1 leptin, 2 adiponectin, 3 resistin, 4 myeloperoxidase (MPO), 5 nitric oxide (NO), 6 superoxide dismutase (SOD), 7 total antioxidant status (TAS), 8 and total oxidant status (TOS). 9 All analyses were performed regarding the manufacturer's instructions using human recombinant standards. Data obtained were reported as concentration for serum and total amount for GCF. All assays were performed at the Department of Biochemistry, Faculty of Medicine, Inonu University.

Statistical analyses
The obtained data were analyzed using a software program (SPSS 22; IBM, Chicago, IL, USA). Whether the parameters showed normal distribution or not was determined by Shapiro-Wilk test. In addition to descriptive statistical analyses, parameters of quantitative data with and without normal distribution were compared between the two groups by Student's t-test and Mann-Whitney U test, respectively. Paired sample t-test and Wilcoxon sign test were used for intra-group comparisons of normally and non-normally distributed parameters, respectively. Continuity (Yates) correction was used to compare qualitative data. A p value < 0.05 was considered statistically significant.

Results
This study was completed with 39 individuals (19 males and 20 females). The flowchart of the study is presented in Fig. 1. There were no significant differences between the two groups in terms of age, gender, and number of teeth (p > 0.05), but the mean BMI of group OB was significantly higher than group C (p < 0.01; Table 1).

Clinical parameters
While there were no significant differences between the groups in terms of clinical parameters and the amount of GCF at baseline (p > 0.05), NSPT provided statistically significant improvements in all parameters in both groups (p < 0.01; Table 2). Likewise, the percentage of sites with shallow (≤ 3 mm), moderate (4-6 mm), and deep (≥ 7 mm) pockets and CAL were similar in both groups at baseline and 3 months after NSPT (p > 0.05), but all of them decreased significantly at 3 months (Table 3).

Biochemical parameters
Serum levels of adipocytokines and OS markers are presented in Fig. 2. Among adipocytokines, TNF-α and leptin levels were higher in group OB at baseline (p < 0.01). NSPT resulted in significant improvements in leptin levels in both groups (p < 0.01, p < 0.05). On the other hand, TNF-α and adiponectin levels improved only in the obese group (p < 0.01, p < 0.05; respectively). There were no statistically significant differences in circulating levels of any adipocytokines between the groups, except leptin, at 3 months. At baseline, only serum TOS level was significantly higher in the obese group (p < 0.01). While NSPT decreased significantly serum NO and TOS levels only in group C (p < 0.01), TAS levels increased in both groups (p < 0.01). TAS and TOS levels were significantly higher and lower in the nonobese group, respectively, after therapy (p < 0.01).
Total amounts of adipocytokines and OS markers in GCF are presented in Fig. 3. While local leptin and SOD levels were significantly higher in group OB at baseline (p < 0.01), TAS and TOS levels were significantly lower and higher, respectively, after therapy (p < 0.05). After NSPT, TNF-α and leptin levels changed significantly only in obese and non-obese groups, respectively (p < 0.05). Periodontal treatment improved all OS markers except NO in both groups, as NO level decreased only in the obese group, when compared to baseline.

Discussion
To the best of our knowledge, this is the first study investigating local and systemic levels of both adipocytokines and OS markers, along with clinical parameters, in response to NSPT. We demonstrated that NSPT contributed periodontally and systemically in both groups. Systemically, obese patients presented a general proinflammatory and oxidative profile at baseline, especially in terms of leptin, TNF-α, and TOS. The clinical improvements accompanied to changes in serum concentrations of the adipocytokines, except resistin, in the obese group. Considerably, obesity appears to modulate the systemic levels of leptin and TAS despite significant changes after periodontal therapy. At the periodontal level, patients with obesity exhibited a proinflammatory status only when regarding leptin at baseline but an oxidative state at 3 months in terms of TAS and TOS.
In the literature, it has been reported that obesity did not appear to impair the success of periodontal therapy but available evidence was variable and inconclusive [15]. Our clinical outcomes suggest that periodontal treatment is effective in reducing local clinical signs of periodontal inflammatory burden and obesity does not have an unfavorable impact on periodontal clinical healing response. These findings are in agreement with previous studies reporting that obesity has no adverse effect on NSPT efficacy [16][17][18][19] despite studies suggesting on the contrary [20][21][22][23]. However, in spite of similar amounts of decrease in both groups, our finding of higher scores of GI in the obese group at 3 months  indicates that obesity increases susceptibility to periodontal inflammation.
Tumor necrosis factor-α is a proinflammatory cytokine that plays a key role in inflammation, immunity, and periodontal disease pathogenesis. It is also released by adipocytes and macrophages in abdominal adipose tissue [2]. TNF-α, originating from adipose tissue, is thought to impair general health by causing insulin resistance, inducing CRP production, and leading to systemic inflammatory status. In accordance with these characteristics, studies have reported increased levels of TNF-α in serum [16,17] and GCF [13,22,24] of obese patients with periodontitis, along with studies in discrepancy [25]. Successful periodontal therapy has been shown to reduce circulating TNF-α levels in patients with periodontitis [26]. Decreased TNF-α levels can increase metabolic control of obesity through insulin resistance. Although both serum and GCF TNF-α levels are thought to be modified by obesity, the findings of the current study suggest that obesity has no significant effect on local TNF-α levels, probably due to the fact that most of the obese individuals in the study were moderately obese. The values of TNF-α, as an important marker of inflammatory status in serum, are compatible with findings reporting that obesity may contribute to the systemic inflammatory state by stimulating cytokine production from adipose tissue.
Contrary to this systemic effect, the local effect on periodontal tissues appears to be limited. Obese patients presented significant decreases in both serum and GCF levels of TNF-α after NSPT. Although obese individuals had higher circulating levels of TNF-α at baseline, they showed more decrease and had similar levels at 3 months. These findings are consistent with previous studies reporting that NSPT reduces circulating and local TNF-α levels in patients with obesity and periodontitis [16][17][18]. Since it has been reported that periodontal inflammation could also increase systemic inflammation [27], the decrease in periodontal inflammation after treatment may have caused a further decrease in serum TNF-α levels of the obese group.
Leptin, being a proinflammatory adipokine, has been determined to play an effective role in local and systemic immuno-inflammatory responses. While serum leptin levels increase in obesity, decrease after weight loss [28]. Recent studies have demonstrated that adipokines such as leptin and adiponectin are also produced in periodontal cells and regulated by periodontopathogenic bacteria [28]. In this study, serum leptin was higher in individuals with obesity both at baseline and 3 months, despite a significant decrease after treatment, confirming prior findings that leptin concentrations are usually increased in patients with obesity presenting or not periodontitis [22,29]. Actually, Fig. 2 Serum levels of adipocytokines and OS markers at baseline and 3 months after NSPT. C: non-obese patients with periodontitis at baseline; C-T: non-obese patients with periodontitis at 3 months; OB: obese patients with periodontitis at baseline; OB-T: obese patients with periodontitis at 3 months. * Significant differences over time within the same group (p < 0.05). ** Significant differences over time within the same group (p < 0.01). † Significant differences between groups at each time point (p < 0.05). † † Significant differences between groups at each time point (p < 0.01) decreased leptin sensitivity and leptin resistance develop in obese patients, causing higher leptin levels [30]. At the periodontal level, both groups presented an increase in the total amount of leptin after NSPT, whereas significantly only in non-obese patients. A possible explanation for the elevated GCF levels of leptin may be the negative correlation between periodontal disease severity and GCF leptin concentration [31]. It has been suggested that gingival inflammation-induced vasodilation might raise leptin levels in serum by increasing the net rate of leptin removal from the gingiva. Findings of this study regarding GCF leptin levels after treatment are in agreement with the studies reporting that periodontally healthy individuals have higher levels of leptin in GCF and gingival tissues than those with periodontitis [31,32].
Adiponectin has anti-inflammatory effects causing suppression of ROS and proinflammatory markers and stimulation of anti-inflammatory markers [33]. Resistin prevents anti-inflammatory effects by stimulating the production of adhesion molecules and other proinflammatory markers. Adiponectin and resistin have been reported to have opposite effects on insulin resistance and inflammation. While plasma levels of adiponectin show a negative correlation with obesity, they have been suggested to decrease in periodontitis and increase again after periodontal treatment [28]. Studies have shown that adiponectin and TNF-α suppress each other's production and function in adipose tissue [34]. On the other hand, although serum resistin levels were proposed to increase in obese patients, there are studies opposite [35]. Zimmermann et al. [13] stated that serum adiponectin and resistin levels were modified by periodontal inflammation independent of obesity. In the present study, serum adiponectin and resistin levels were similar in both groups at baseline concordant with studies reporting likewise [13,22,25]. At 3 months, we observed a significant increase in serum adiponectin level only in the obese group. The decrease in TNF-α and increase in adiponectin levels after therapy support the argument that these two adipokines block each other. In obesity, increased TNF-α and resistin with decreased adiponectin levels trigger the proinflammatory process. The increase in adiponectin level at 3 months after NSPT indicates a positive immunological response in obese individuals. Current findings show that there is no additional inflammatory burden due to systemic levels of adiponectin and resistin in the obese group.
Oxidative stress plays a significant role in the pathogenesis of many diseases including periodontitis and obesity. Studies suggest that obesity may contribute to the destruction of periodontal tissues by affecting local oxidative parameters [36]. Myeloperoxidase is a ROS-producing bactericidal enzyme and MPO activity with neutrophil infiltration has been shown to increase in obese individuals [37]. Moreover, increased MPO activity in periodontally diseased areas and decreased activity after treatment supports the role of MPO in destructive periodontal diseases [38]. In this study, there were significant decreases in GCF MPO levels in both groups after treatment. We can conclude that neutrophil migration into the pocket was similar in both groups and periodontal therapy was effective in reducing the inflammatory condition with local MPO levels. Fig. 3 Total amounts of adipocytokines and OS markers in GCF at baseline and 3 months after NSPT. C: non-obese patients with periodontitis at baseline; C-T: non-obese patients with periodontitis at 3 months; OB: obese patients with periodontitis at baseline; OB-T: obese patients with periodontitis at 3 months. * Significant differences over time within the same group (p < 0.05). ** Significant differences over time within the same group (p < 0.01). † Significant differences between groups at each time point (p < 0.05). † † Significant differences between groups at each time point (p < 0.01) Cells have developed antioxidant defense systems to protect themselves against OS. SOD is the most important antioxidant enzyme that catalyzes the conversion of superoxide radicals to less reactive forms and protects the cells against ROS [39]. In obesity, macrophage infiltration in adipose tissue leads to an increase in free radicals as a result of the raised release of inflammatory cytokines, causing OS. Although antioxidant activity has been reported to be significantly lower in obese individuals, opposing results have been suggested too [40]. The only study in the literature, evaluating SOD in obese patients with periodontitis after NSPT, reported an increase in SOD activity after therapy along with weight loss [41]. In our study, GCF SOD level was significantly higher in obese group at baseline. SOD activity might have increased as an adaptive response to excess superoxide production because of chronic inflammation and OS in obesity [42]. Presence of SOD in human periodontal ligament and the stimulation of superoxide release from gingival fibroblasts by bacterial LPS during inflammation may be another reason for increased SOD activity [43]. Local SOD levels significantly increased after treatment in both groups, indicating that periodontal therapy contributed to local antioxidant defense with improvement in clinical parameters. Current findings confirm the studies reporting that SOD levels were higher in periodontally healthy individuals and anti-oxidant defense increased after periodontal treatment [43,44].
Nitric oxide is a free radical involved in a wide range of processes such as vasodilatation and inflammation and its synthesis is regulated by nitric oxide synthase (NOS). Although NO synthesis has been reported to be an indicator of abdominal obesity, OS, and inflammation, data in the literature are contradictory [45]. While increased LDL/ HDL ratio and insulin resistance, decreased adiponectin concentration in obesity reduces eNOS activity, and inflammation causes a rise in iNOS activity. As a result of the upregulation of iNOS, which stimulates metalloproteinases, NO levels are elevated in periodontitis and periimplantitis [46]. It has been reported that increased expression of iNOS in inflamed periodontal tissues decreased after NSPT [47]. Since NO is not stable, NO level was measured via its stable end products; nitrite and nitrate. In the current study, serum and GCF NO levels decreased significantly in obese and non-obese groups, respectively, 3 months after treatment. The fact that NO production results from complex interactions between NOS activity and factors regulating its expression and may change throughout the obesity process supports our findings.
Total anti-oxidant status is a biochemical parameter that reflects all of the antioxidants in the examined biological samples, including those yet to be discovered. Studies have suggested that obese individuals had significantly lower TAS values than non-obese subjects, TAS was negatively correlated with WC and BMI, and especially abdominal obesity was associated with increased OS and decreased TAS [42,48]. However, there are also studies reporting contrarily due to an antioxidant-rich diet or an effective response against the oxidant system [49]. Atabay et al. [50] detected lower TAS levels in obese patients with periodontitis but they did not perform any treatment. In this study, both local and peripheral TAS were significantly increased after NSPT, concordant with the reports suggesting increased levels following successful periodontal therapy [51,52].
While TAS is used to assess the general status of antioxidants, TOS is similarly used to determine the level of oxidants. TOS measurement is a practical and reliable method that reflects the total oxidative effect resulting from the possible interactions of different oxidant molecules with each other besides their oxidative activities. It has been suggested that oxidative parameters were higher in obese individuals and TOS decreased as a result of fat and weight loss after diet and exercise [53]. Knaś et al. [42] reported lower TAS and higher TOS in morbidly obese and reported that their levels increased and decreased, respectively, after bariatric surgery. In our study, serum TOS was found to be higher in the obese group, suggesting that oxidative status was modified by obesity systemically, but not locally, contrary to the study reporting higher GCF TOS values in obese women [54]. Although GCF and serum TOS values have been shown to decrease after periodontal treatment in the literature [39,55], GCF TOS decreased in both groups, but serum TOS only in the non-obese group. This appears to be consistent with our baseline serum TOS values modified by obesity. Since there are just two studies evaluating OS after NSPT in obese individuals with periodontitis [41,56], and a single study examining only SOD among the markers we examined [41], we could not compare our results completely.
The major strength of this study is that this is, to our knowledge, the first study to evaluate the effects of NSPT on systemic and local adipocytokines and OS markers concurrently in obese patients with periodontitis. The exclusion of patients with diabetes, smokers, and overweight individuals makes our findings stronger. While defining patients as obese or non-obese, we considered both BMI and WC as indicators of abdominal adiposity and body fat distribution. We also repeated anthropometric measurements at the end of the study to ensure that the patients' obesity status did not change. On the other hand, the lack of periodontally healthy groups, which could supply information about the inflammation induced by periodontitis, is a limitation of this study. Additionally, most of the patients in the obese group were mildly/moderately obese so these findings cannot reflect the status of those classified as morbidly obese. The association between periodontitis and obesity might be more pronounced as the status of obesity worsens.

Conclusion
In conclusion, our results suggest that obesity may modulate the inflammatory and oxidative profile of patients with periodontitis by increasing the levels of adipocytokines and OS markers, both systemically and locally. Although periodontal therapy improved clinical periodontal parameters, these improvements were not compatible with the changes in local and circulating levels of some inflammatory and oxidative biomarkers. Higher proinflammatory and oxidant status in patients with obesity may contribute to their being more prone to periodontitis.