CC BY
55
A PROGNOSTIC MODEL FOR THE PREDICTION OF GENERALIZED CHRONIC PERIODONTITIS IN PATIENTS WITH METABOLIC SYNDROME
Petrukhina NB12 Zorina OA1-2' Shikh EV1, Kartysheva EV1«, Kudryavtsev AV2
1 Sechenov First Moscow State Medical University, Moscow, Russia
2 Central Research Institute of Dentistry and Maxillofacial Surgery, Moscow, Russia
The pathogenesis of periodontitis involves a complex inflammatory cascade initiated by biofilm bacteria. The susceptibility to or the risk of developing the disease is determined by the body's response to the invasion, specifically, by the strength of the inflammatory response and the differential activation of immune pathways. In this paper, we propose a model for predicting the risk of severe chronic generalized periodontitis (GCP) in patients with metabolic syndrome based on the levels of tumor necrosis factor alpha (TNF-a) in the periodontal pocket exudate. The analysis of oral cavity cytokine profiles conducted in 537 patients with GCP and comorbid metabolic syndrome showed that increased TNF-a correlated with the severity of GCP: higher levels of TNF-a were observed in patients whose condition was more severe. The prognostic model built in Statistica. 10 allowed us to use TNF-a as a prognostic criterium for GCP severity. We determined the cut-off point above which a high risk of severe GCP can be concluded with 91.2% sensitivity and 70.8% specificity. The spreadsheet in Microsoft Excel 2010 automatically computed the risk of severe GCP from a patient's TNF-a concentrations in the PP, which makes the model convenient for routine clinical use in dentistry.
Keywords: periodontitis, metabolic syndrome, cytokines, tumor necrosis factor, TNF, prognostic model
Author contribution: Petrukhina NB and Shikh EV conceived the study, analyzed and interpreted the obtained data; Zorina OA analyzed and interpreted the obtained data; Kartysheva EV collected the samples, analyzed the literature and interpreted the obtained data; Kudryavtsev AV prepared the manuscript draft and analyze the literature.
Compliance with ethical standards: the study was approved by the Ethics Committee of I.M.Sechenov First Moscow State Medical University (Protocol No. 10-15 dated November 18, 2015). All patients gave informed consent to participate.
[X] Correspondence should be addressed: Ekaterina V. Kartysheva,
Bolshaya Pirogovskaya 2, bld. 4, Moscow, 119991; 89057676464@mail.ru
Received: 24.09.2018 Accepted: 31.03.2019 Published online: 14.04.2019
DOI: 10.24075/brsmu.2019.026
ПРОГНОСТИЧЕСКАЯ МОДЕЛЬ ДЛЯ ОЦЕНКИ ХРОНИЧЕСКОГО ГЕНЕРАЛИЗОВАННОГО ПАРОДОНТИТА У ПАЦИЕНТОВ С МЕТАБОЛИЧЕСКИМ СИНДРОМОМ
Н. Б. Петрухина1,2, О. А. Зорина1,2, Е. В. Ших1, Е. В. Картышева1 А. В. Кудрявцев2
1 Первый Московский государственный медицинский университет имени И. М. Сеченова (Сеченовский Университет), Москва, Россия
2 Центральный научно-исследовательский институт стоматологии и челюстно-лицевой хирургии, Москва, Россия
Патогенез пародонтита включает сложный иммунный воспалительный каскад, который инициируется бактериями биопленки, а восприимчивость или вероятность развития заболевания определяется реакцией организма человека, в частности, величиной воспалительного ответа и дифференциальной активацией иммунных путей. Цель исследования — разработать прогностическую модель для оценки риска развития тяжелой степени хронического генерализованного пародонтита в зависимости от содержания фактора некроза опухоли-a (ФНО-a) в экссудате пародонтального кармана (ПК) пациента. При клинико-инструментальном обследовании 537 пациентов с хроническим генерализованным пародонтитом и метаболическим синдромом установлено, что уровень повышения ФНО-a в содержимом ПК коррелировал со степенью тяжести хронического генерализованного пародонтита (ХГП): более высокие значения цитокина соответствовали более тяжелой степени. Разработанная в программе Statistica.10 прогностическая модель дала возможность использовать уровень ФНО-a в содержимом ПК пациента в качестве прогностического критерия течения ХГП. Определено критическое значение, при превышении которого с диагностической чувствительностью 91,2% и специфичностью 70,8% можно заключить о высоком риске развития тяжелой степени ХГП. Созданное окно в программе Microsoft Excel 2010 позволяет автоматически рассчитывать риск развития тяжелой степени ХГП в зависимости от индивидуального значения концентрации ФНО-a в содержимом ПК пациента, что делает данную модель удобной для применения врачами-стоматологами.
Ключевые слова: пародонтит, метаболический синдром, цитокины, фактор некроза опухоли, ФНО, прогностическая модель
Информация о вкладе авторов: Н. Б. Петрухина; Е. В. Ших — планирование исследования, анализ и интерпретация данных; О. А. Зорина — анализ и интерпретация данных; Е. В. Картышева — сбор данных, анализ и интерпретация данных; А. В. Кудрявцев — черновик рукописи, анализ литературы.
Соблюдение этических стандартов: этический комитет Первого МГМУ им. И. М. Сеченова Сеченовский Университет) (протокол № 10-15 от 18 ноября 2015 г); все пациенты подписали добровольное согласие на участие в исследовании.
Рк] Для корреспонденции: Екатерина Владимировна Картышева,
ул. Большая Пироговская, д. 2, стр. 4, г Москва, 119991; 89057676464@mail.ru
Статья получена: 24.09.2018 Статья принята к печати: 31.03.2019 Опубликована онлайн: 14.04.2019 DOI: 10.24075/vrgmu.2019.026
Tumor necrosis factor alpha (TNF-a) is one of the early proinflammatory cytokines that plays a key role in periodontal tissue destruction [1]. Clinical studies have demonstrated that elevated TNF-a is a risk factor for the progression of periodontal diseases. TNF-a mediates periodontal tissue destruction via at least two different pathways. Firstly, it stimulates production of osteoclasts that cause alveolar bone resorption [2, 3]. Secondly, TNF-a promotes the body's early immune response to periodontal pathogens and regulates synthesis of matrix
metalloproteinases (MMP) capable of damaging connective tissue. Besides, it is reported that TNF-a levels are increased systemically in patients with obesity or metabolic syndrome [4]. Adipose tissue cells secrete TNF-a; therefore, accumulation of excess fat leads to chronic systemic inflammation [5, 6]. It has been shown that TNF-a levels also correlate with insulin resistance [7]. TNF-a is a paracrine mediator: its local activity is aimed at reducing sensitivity of adipocytes to insulin [8]. There is a reciprocal connection between periodontal disease and
metabolic syndrome. The severity of systemic inflammation in patients with metabolic syndrome can affect local inflammation in the periodontium, while the products of periodontal inflammation can stimulate secretion of systemic cytokines. In this study we propose a prognostic model for predicting the risk of severe generalized chronic periodontitis based on TNF-a concentrations in the exudate from a periodontal pocket (PP).
METHODS
We examined 537 patients (243 females and 294 males; 45.25% vs. 54.75%, respectively) aged 35 to 65 years with clinically diagnosed generalized chronic periodontitis (GCP) and metabolic syndrome. The patients were distributed into 3 age groups: group 1 included patients aged 35-44 years, with the mean age of 41.7 ± 2.1; group 2 consisted of patients aged 45-54 years, with the mean age of 52.2 ± 1.2; group 3 comprised individuals aged 55-65 years, with the mean age of 63.4 ± 1.1. Our study included patients of both sexes aged 35 to 65 years, with clinically diagnosed GCP, comorbid metabolic syndrome and a body mass index > 25 kg/m2, who gave written informed consent to participate. The following exclusion criteria were applied: age under 35 years; hematologic disorders; diseases of the central nervous system, both congenital and acquired; malignancies (cancers, sarcomas); decompensated chronic conditions (myocardial infarction, systemic thromboembolism); pregnancy.
Exudate samples were collected onto filter paper strips introduced into the PP for 30 s. Then, the strips were transferred into Eppendorf tubes containing 1 ml of sterile normal saline and left there for 40 min. After that, the strips were taken out with tweezers, and the content of the Eppendorf tubes was
analyzed. TNF-a levels were measured using ELISA kits by BIOSOURCE (Europe S. A.; Belgium); spectrophotometry was done by a microplate reader at 450 nm wavelength. Cytokine concentrations were determined from a standard curve and expressed as pg/ml.
The next step was to create a prognostic model for predicting the risk of developing severe GCP based on TNF-a concentrations. A primary data matrix was generated in Statistica.10 (StatSoft; USA). Model coefficients were calculated in the output spreadsheet and included into the mathematical expression. Then, a ROC curve was constructed and a cut-off point was determined. The cut-off point allows using the model for practical tasks: new data can be assigned to one of the 2 classes depending on their position relative to the cut-off point. Besides, we applied the ROC-curve analysis to assess the diagnostic efficacy of our model by calculating the AUC value (Area Under Curve) using a trapezoidal rule.
RESULTS
Based on the obtained TNF-a concentrations (Table 1), a prognostic model was built for predicting the risk of developing severe GCP.
The mathematical expression below can be used to calculate the risk of severe periodontal tissue destruction based on the TNF-a concentrations in the PP. The measured TNF-a concentrations should be plugged into the following formula:
W = -3.2 + 1.2 • log10 (Y),
where W is a risk of developing severe GCP calculated from the cytokine profile of the oral cavity and Y is a TNF-a concentration in the PP expressed as pg/ml.
Table 1. Levels of the key proinflammatory cytokine TNF-a in the PP exudate in patients with GCP and comorbid metabolic syndrome
Parameter 35-44 years 45-54 years 55-65 years
Sex М F М F М F
Mild GCP
TNF-a (pg/ml) 576.80 ± 19.49 584.96 ± 21.54 611.78 ± 21.67 634.57 ± 23.5 645.67 ± 23.7 678.45 ± 24.9
Moderate GCP
TNF-a (pg/ml) 845.44 ± 32.76 876.5 ± 33.7 848.34 ± 24.5 998.56 ± 21.5* 945.81 ± 32.33 1045.33 ± 31.56
Severe GCP
TNF-a (pg/ml) 878.93 ± 32.11 911.23 ± 31.67 905.78 ± 35.6 1145.87 ± 35.11* 1234.56 ± 33.17 1341.54 ± 33.98
Note: * — shows that the difference is significant.
TNF-a (pg/ml)
Fig. 1. The graph shoes the relationship between the risk of developing severe GCP and the TNF-a concentrations in the PP
20 -
0 h' ..,>.,.»... t ....... t
0 20 40 60 80 100 Specificity (%)
Fig. 2. The ROC-curve represents a relationship between the diagnostic sensitivity and specificity in predicting the risk of severe GCP from the cytokine profile of the oral cavity
Fig. 1 is the graphic representation of the relationship between the risk of developing severe GCP and the TNF-a concentrations in the PP. The risk for severe GCP increases as TNF-a levels grow in the PP.
The risk W for developing severe GCP was calculated for each study participant from TNF-a concentrations in the PP. Then, the ROC analysis was conducted to determine a critical (cut-off) value for W (0.3), above which a high risk for severe GCP could be predicted with maximum specificity and sensitivity.
W > 0.3 means that the risk for severe GCP is high; W < 0.3 means it is low. The diagnostic sensitivity of the method is 91.2%, whereas its specificity is 70.8%.
Fig. 2 features a ROC curve for different values of the prognostic coefficient W. Table 2 shows sensitivity and specificity of the method at which W = 0.3 had the highest sensitivity and specificity.
The AUC value of 0.862 ± 0.05 (z = 7.3; p < 0.001) and the confidence interval of 0.765-0.959 suggest that W has a high diagnostic significance in predicting severe GCP based on the cytokine profile of the oral cavity.
W was computed in Microsoft Excel 2010; individual TNF-a concentrations were entered into the highlighted cell (Fig. 3).
DISCUSSION
TNF-a plays a key role in the pathogenesis of periodontal diseases. When bacterial lipopolysaccharides permeate periodontal tissue, macrophages assisted by CD14-
lymphocytes activate a number of innate and adaptive immunity mechanisms through specific receptors. An inadequately strong immune response leads to chronic inflammation and periodontal tissue destruction [9, 10]. Prostaglandin E2, IL1P and TNF-a are key proinflammatory mediators that activate tissue metalloproteinases and thereby stimulate bone resorption by osteoclasts and induce damage to the periodontium [11]. A number of nonimmune periodontal cells, such as epithelial cells and fibroblasts, can recognize and respond to proinflammatory IL1 p and TNF-a. Tissue metalloproteinases produced by neutrophils, macrophages, fibroblasts, and osteoclasts promote proteolysis of collagen, gelatin and elastin, destroying the connective tissue components of tooth-supporting structures. Among the members of the TNF superfamily are osteotropic factors, such as the receptor activator of NF-kB ligand (RANKL) and RANK themselves that are synthesized by osteoclasts and promote bone resorption [12]. The binding of the RANK ligand to the RANK receptor is accompanied by a fusion of a few precursor cells into a mature multinucleated osteoclast that immediately starts to destroy bone tissue (Fig. 4).
In light of this, the study of TNF-a levels in the PP of patients opens new possibilities for predicting the severity of GCP.
CONCLUSIONS
We have found that elevated TNF-a concentrations in the PP correlate with the severity of GCP in patients with metabolic syndrome: higher TNF-a levels are associated with a more
Table 2. Sensitivity and specificity of the method at different values of the prognostic coefficient W used to predict the risk of severe GCP from the cytokine profile of the oral cavity
Parameter W Diagnostic sensitivity CI for DSn Diagnostic. specificity CI for DSp
> 0.29 94.44 81.3 - 99.3 62.5 40.6 - 81.2
> 0.3* 91.67 77.5 - 98.2 70.83 48.9 - 87.4
> 0.32 86.11 70.5 - 95.3 70.83 48.9 - 87.4
> 0.33 83.33 67.2 - 93.6 70.83 48.9 - 87.4
> 0.34 80.56 64.0 - 91.8 70.83 48.9 - 87.4
> 0.35 72.22 54.8 - 85.8 70.83 48.9 - 87.4
> 0.36 69.44 51.9 - 83.7 75 53.3 - 90.2
> 0.38 66.67 49.0 - 81.4 79.17 57.8 - 92.9
> 0.4 63.89 46.2 - 79.2 83.33 62.6 - 95.3
> 0.41 61.11 43.5 - 76.9 87.5 67.6 - 97.3
> 0.45 61.11 43.5 - 76.9 91.67 73.0 - 99.0
> 0.46 55.56 38.1 - 72.1 91.67 73.0 - 99.0
> 0.5 44.44 27.9 - 61.9 91.67 73.0 - 99.0
> 0.51 36.11 20.8 - 53.8 91.67 73.0 - 99.0
> 0.52 27.78 14.2 - 45.2 100 85.8 - 100.0
Note: * — marks the cut-off point.
TNF-a concentrations in a patient's periodontal pocket Cut-off point (calculated using ROC analysis) 0.3
Diagnostic sensitivity 91.7
1DC0 pg/ml 0.50 Diagnostic specificity 70.8
Z statistic 7.3
The risk for severe GCP P < 0.001
Area under curve (AUC) 0.862
Fig. 3. A spreadsheet In Microsoft Excel 2010 used for automatic calculations of the risk for severe GCP based on the oral cavity cytokine profile
LPS, MAMPs
Stromal cells, _ .. . _ ..
f I ... . ' . T cells/ B cells opg
fibroblasts, osteoblasts OPG
Bait" receptor
Osteoclast activation
IL1 Activation
Monocyte Precursor cell
Bone resorption
Fig. 4. Integration of proinflammatory and osteotropic factors in bone resorption
severe course of the disease. The proposed prognostic model based on the TNF-a concentrations in the PP is a promising and informative noninvasive method that can be used to predict
the progression of the disease. The advantages of our model include low costs, availability, fast results, and the simplicity of use, which is crucial for routine dental practice.
References
1. Graves D, Cochran D. The contribution of interleukin-1 and tumor necrosis factor to periodontal tissue destruction. J Periodontol. 2003; 74 (3): 391-401.
2. Aganov DS, Tyrenko VV, Cygan EN, Toporkov MM, Bologov SG. Rol' citokinovoj sistemy RANKL/RANK/OPG v regulyacii mineral'nogo obmena kostnoj tkani. Geny i kletki. 2014; (4): 50-2. Russian.
3. Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science. 1993; 259 (5091): 87-91.
4. Petrukhina NB, Rabinovich IM, Zorina OA. Narushenie mikrobiocenoza rta u pacientov s metabolicheskim sindromom (chast'1). Institut stomatologii. 2014; 1 (62): 54-7. Russian.
5. Hivert MF, Sullivan LM, Fox CS, Nathan DM, D'Agostino RB Sr, Wilson PW et al. Associations of adiponectin, resistin, and tumor necrosis factor-alpha with insulin resistance. J Clin Endocrinol Metab. 2008; 93 (8): 3165-72.
6. Hotamisligil GS, Arner P, Caro JF, Atkinson RL, Spiegelman BM. Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J Clin Invest. 2005; 95 (5): 2409-15.
7. Gregor MF, Hotamisligil GS. Inflammatory mechanisms in obesity. Annu Rev Immunol. 2011; (29): 415-45.
8. Kobayashi K, Takahashi N, Jimi E, Udagawa N, Takami M, Kotake S. et al. Tumor necrosis factor alpha stimulates osteoclast differentiation by a mechanism independent of the ODF/RANKL-RANK interaction. J Exp Med. 2000; 191 (2): 275-86.
9. Lau DC, Dhillon B, Yan H, Szmitko PE, Verma S. Adipokines: molecular links between obesity and atheroslcerosis. Am J Physiol Heart Circ Physiol. 2005; 288 (5): 2031-41.
10. Zorina OA, Rabinovich IM, Petrukhina NB, Kudryavtseva EV. Sravnitel'nye rezul'taty antropometrii i bioimpedansnogo issledovaniya u pacientov s hronicheskim generalizovannym parodontitom i metabolicheskim sindromom. Stomatologiya. 2016; 95 (2): 91-2. Russian.
11. Barer G, Grigoryan S, Postnova N. Rol' interferona i drugih citokinov v vozniknovenii i razvitii zabolevanij parodonta. Cathedra. 2006; 5 (3): 54-60. Russian.
12. Sagalovsky S, Schönert M. RANKL-RANK-OPG system and bone remodeling: a new approach on the treatment of osteoporosis. Clin Exptl Pathol. 2011; 2 (10): 146-53.
Литература
1. Graves D, Cochran D. The contribution of interleukin-1 and tumor necrosis factor to periodontal tissue destruction. J Periodontol. 2003; 74 (3): 391-401.
2. Аганов Д. С., Тыренко В. В., Цыган Е. Н., Топорков М. М., Бологов С. Г. Роль цитокиновой системы RANKL/RANK/OPG в регуляции минерального обмена костной ткани. Гены и клетки. 2014; (4): 50-2.
3. Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science. 1993; 259 (5091): 87-91.
4. Петрухина Н. Б., Рабинович И. М., Зорина О. А. Нарушение микробиоценоза рта у пациентов с метаболическим синдромом (часть 1). Институт стоматологии. 2014; 1 (62): 54-7.
5. Hivert MF, Sullivan LM, Fox CS, Nathan DM, D'Agostino RB Sr, Wilson PW et al. Associations of adiponectin, resistin, and tumor necrosis factor-alpha with insulin resistance. J Clin Endocrinol Metab. 2008; 93 (8): 3165-72.
6. Hotamisligil GS, Arner P, Caro JF, Atkinson RL, Spiegelman BM. Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J Clin Invest. 2005; 95 (5): 2409-15.
7. Gregor MF, Hotamisligil GS. Inflammatory mechanisms in obesity. Annu Rev Immunol. 2011; (29): 415-45.
8. Kobayashi K, Takahashi N, Jimi E, Udagawa N, Takami M, Kotake S. et al. Tumor necrosis factor alpha stimulates osteoclast differentiation by a mechanism independent of the ODF/RANKL-
RANK interaction. J Exp Med. 2000; 191 (2): 275-86.
9. Lau DC, Dhillon B, Yan H, Szmitko PE, Verma S. Adipokines: molecular links between obesity and atheroslcerosis. Am J Physiol Heart Circ Physiol. 2005; 288 (5): 2031-41.
10. Зорина О. А., Рабинович И. М., Петрухина Н. Б., Кудрявцева Е. В. Сравнительные результаты антропометрии и биоимпедансного исследования у пациентов с хроническим генерализованным пародонтитом и метаболическим синдромом. Стоматология. 2016; 95 (2): 91-2.
11. Барер Г., Григорян С., Постнова Н. Роль интерферона и других цитокинов в возникновении и развитии заболеваний пародонта. Cathedra. 2006; 5 (3): 54-60.
12. Sagalovsky S, Schonert M. RANKL-RANK-OPG system and bone remodeling: a new approach on the treatment of osteoporosis. Clin Exptl Pathol. 2011; 2 (10): 146-153.
