Buduneli N, Kinane DF: Host-derived diagnostic markers related to soft tissue destruction and bone degradation in periodontitis. J Clin Periodontol 2011; 38 (Suppl. 11): 85–105. doi: 10.1111/j.1600-051X.2010.01670.x.
Background: A major challenge in clinical periodontics is to find a reliable molecular marker of periodontal tissue destruction with high sensitivity, specificity and utility.
Objectives: The aim of this systematic review is to evaluate available literature on ‘the utility of molecular markers of soft and hard periodontal tissue destruction’.
Materials and Methods: Based on the focused question, ‘What is the utility of molecular markers of soft and hard periodontal tissue destruction’, an electronic and manual search was conducted for human studies presenting clinical data for the potential of molecular markers of tissue destruction in biofluids; gingival crevicular fluid (GCF), saliva, and serum.
Results: Papers fulfilling the inclusion criteria were selected. All relevant data from the selected papers were extracted and recorded in separate tables for molecules in GCF, saliva, and serum.
Conclusion: Within the defined limits of the Problem/Population, Intervention, Comparison, Outcome, the present analysis reveals that (a) no single or combination of markers exists that can disclose periodontal tissue destruction adequately; (b) while the most fruitful source of biomarkers for periodontal destruction appears to be in molecules tightly related to bone and soft tissue destruction, this remains to be objectively demonstrated. Currently, clinical measurements are still the most reliable.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Periodontitis is one of the most common oral diseases and is characterized by gingival inflammation and alveolar bone resorption (Savage et al. 2009). More than 500 different bacterial species are able to colonize the oral biofilm and up to 150 different species of bacteria are possible in any individual's subgingival plaque. According to a report by the World Health Organisation, severe periodontitis leading to tooth loss was found in 5–15% of most populations worldwide (Armitage 2004). Hence, it can be considered among the prevalent and important global health problems in terms of quality of life.
A diagnostic tool should provide pertinent information to aid differential diagnosis, screening, presence, location, severity or staging and prognosis of a disease. At present, periodontitis is diagnosed almost entirely on the basis of an array of clinical measurements including probing depth (PD), clinical attachment level (CAL), bleeding on probing (BOP), plaque index (PI) recordings and radiographical findings. However, PD and CAL measurements by periodontal probes and radiographic bone levels, provide information about past periodontal tissue destruction and do not elucidate the current state of the disease activity nor predict the future. Therefore, one of the major challenges in the field of periodontology is to discover a method of predicting the future of periodontal disease or at least to declare the current state of disease activity. Thus, objective and ideal diagnostic methods for periodontal diagnosis are still being sought. The ideal periodontal diagnostic method should be able firstly to screen susceptible subjects in the general population, secondly to differentiate active and inactive sites, thirdly to predict future tissue destruction in particular individuals and sites, and finally to monitor the response to periodontal therapy.
Periodontitis is described as a multi-factorial irreversible and cumulative condition, initiated and propagated by bacteria and host factors (Kinane 2001). Given the complex nature of periodontitis, it is unlikely that one single clinical or laboratory examination can address all issues concerning diagnosis, classification, and prognosis (Van der Velden 2005, Xiang et al. 2010). Moreover, molecular markers of bone resorption have advantages as well as disadvantages as they relate to specificity for bone, ease of detection, pre-analytic stability, and availability of sensitive and specific assays for detection (Koka et al. 2006). Proteins derived from inflamed host tissue and pathogenic bacteria have the potential of being used as markers of periodontitis (Özmeriç 2004).
Numerous molecules in the oral fluids namely; gingival crevicular fluid (GCF) and saliva, as well as molecules in the blood circulation; serum or plasma have been investigated so far in an attempt to provide a sensitive and specific marker for periodontal tissue destruction. GCF and saliva are particularly promising as they can easily be obtained non-invasively and with minimal discomfort to the patient and consist of both locally synthesized and systemically derived molecules. Thus, evaluation of various biologically specific proteins or markers in oral fluids by using immunologic or biochemical methods may provide information on the events going on in the periodontal microenvironment (Kinney et al. 2007).
The immune and inflammatory responses are critical to understanding the pathogenesis of periodontal diseases and they are orchestrated by a number of host-related factors, either intrinsic or induced (Taubman et al. 2005). Under normal physiologic conditions, there is a balance between bone formation and bone resorption. Bone homoeostasis is maintained as long as this balance is preserved. When the structural integrity and/or calcium metabolism is altered, this balance is lost towards either increased bone formation or towards bone resorption. Inflammatory conditions like periodontal diseases and rheumatoid arthritis, and metabolic conditions like osteoporosis are examples of an altered balance between bone formation and bone resorption.
The threshold for periodontitis progression is based upon extensively documented evidence within the periodontal literature. Many epidemiological studies apply a threshold for loss of attachment of changes of 3 mm as the definition of disease progression. The threshold is set at the level of two teeth to minimize the risk of including cases of progression arising because of reasons other than periodontitis and/or measurement error. Presence of 2 teeth demonstrating a longitudinal loss of proximal attachment of 3 mm has been proposed as the criteria for case definition of periodontitis progression (Tonetti & Claffey 2005). In situations where serial proximal attachment level measurements are not available, longitudinal radiographic bone loss of 2 mm at 2 teeth may be used as a substitute.
The criteria for definition of progressing sites are as stated by the working group in the Fifth European Workshop in Periodontology, in 2005as follows:
- •The exposure of interest, along with relevant covariates including age, is used in a multi-factorial model, using the ‘‘case’’ as the dependent variable. Adjusted odds ratios and/or relative risk estimates as well as 95% confidence limits should be reported.
- •The impact of the putative risk factor on the extent of periodontitis progression (percentage of teeth affected by disease progression) is examined to investigate a dose–response effect, using a new multi-variate model.
- •Additional models may be developed as needed to test specific hypotheses.
Clinical examination and the clinical periodontal measurements can provide information on the disease activity, only if these measurements are repeated at two time-points. However, knowing the disease activity state might be critical to clinical decision making at one single time point and clinical evaluation cannot fulfil this requirement. This deficiency has spawned dozens of studies with as yet no definitive answer.
There are also well-accepted risk factors such as diabetes and smoking to be considered when determining the diagnosis as well as prognosis of a particular case. Cigarette smoking represents a risk factor for progression of periodontitis, the effect of which may be dose related. Indeed, apart from the plaque biofilm, it is the major environmental risk factor. Heavy smokers should be considered as high-risk individuals for disease progression. The clinical implications for this are that smokers should be identified during patient examination and efforts should be made to modify this behavioural risk factor. Furthermore, smoking or molecules related to smoking such as blood cotinine induced by smoking should be considered as important risk markers of periodontal disease that are relevant to the assessment of prognosis (Calsina et al. 2002, Tang et al. 2009).
The aim of the present review was to provide a systematic review of the state of evidence on whether current periodontal tissue destruction can be disclosed accurately utilizing chemical analysis of various molecules in the biological fluids.
Materials and Methods
A literature search of the last thirty years was performed using the ISI and PubMed database from 1980 to 15 June 2010, with the following search strategy: (“periodontitis” OR “periodontal disease”) AND (“progression” OR “activity”) AND (“saliva” OR “gingival crevicular fluid” OR “serum” OR “plasma”). The search was limited to the English language. In vitro studies on cell cultures, experimental studies on animal models, polymorphism studies, studies particularly investigating possible role of various therapeutic agents such as subantimicrobial dose doxycycline, anti-inflammatory agents, or dietary supplements, as well as studies comprising only patients with particular systemic diseases such as diabetes or rheumatoid arthritis and studies focused only on smoking were excluded from the present review. Titles and abstracts were screened and full text of publications was obtained for the selected articles. All levels of evidence were included. In addition, the reference lists of review papers were hand searched. To be eligible for inclusion in this systematic review, studies had to meet the following criteria: (1) original investigations; (2) studies conducted within a human population; (3) studies having a systemically healthy chronic periodontitis group; (4) studies correlating the biochemical findings in biofluids with clinical periodontal parameters: PD, CAL, and BOP; (5) preferentially follow-up and/or intervention studies, but cross-sectional studies were also included in the review.
The primary outcome variables investigated were PD, CAL, and incidence of BOP.
Clinical methods for periodontal diagnosis
Measuring PD and CAL by a calibrated periodontal probe and assessing gingival inflammation by gingival index (GI) (Löe & Silness 1963) and/or a bleeding index such as BOP within a certain time after probing or papilla bleeding index (PBI) (Saxer & Mühlemann 1975) are still the most reliable clinical periodontal parameters for periodontal diagnosis. Furthermore, PI (Silness & Löe 1964) provides information on the major local aetiological factor; microbial dental plaque. Presence of BOP is still the most reliable clinical finding indicating periodontal disease activity. However, the absence of bleeding is much more valuable in terms of being a highly specific negative predictor of periodontal disease activity (Lang et al. 1986, 1990). This helps us exclude healthy patients but of course does not help us truly focus on the really at risk subjects as if we were to address all patients with BOP we would be overwhelmed with false positives as this approach would be far too sensitive to be useful as a screening technique. Sensitivity is the ability of a diagnostic test to identify the target molecule when this is truly present. Specificity is the probability of a diagnostic test being negative when the target molecule is truly absent. Misleading test results occur when the test is positive and the disease is absent (false positive) or when the test is negative and the disease is present (false negative). An ideal diagnostic test should have sensitivity and specificity values approaching 100%. But unfortunately this is never the case.
Source for samples of the different diagnostic tests
GCF is a transudate originating from the gingival plexus of blood vessels in the gingival connective tissue, close to the epithelium lining of the dentogingival space. It also collects resident host cells and microorganisms in the microbial dental plaque as well as their cellular products. GCF provides an accurate representation of tissue and serum concentrations of inflammatory mediators (Giannobile 1997, Champagne et al. 2003, Armitage 2004). Up to now, at least 90 different components in GCF have been evaluated as possible biomarkers for diagnosis of periodontal disease (Loos & Tjoa 2005). These markers can be divided into three major groups: host-derived enzymes and their inhibitors, inflammatory mediators and host-response modifiers, and byproducts of tissue breakdown (Lamster & Ahlo 2007).
A substantial number of studies throughout the 1980s and the 1990s explored the predictive ability of GCF components for identification of progressive periodontal lesions (Heitz-Mayfield 2005). While individual GCF components produced positive predictive values that were superior to individual clinical measures (Chapple et al. 1999), these studies focused largely on the prediction of periodontitis at the site level rather than the identification of high-risk groups and individuals. Data from the selected studies investigating potential biomarkers in GCF in relation to periodontal disease are outlined in Table 1. An early multi-centre study by Lamster et al. (1995) did examine the predictive value of β-glucuronidase (βG) at the patient level, in a population of predominantly recall patients, and demonstrated that subjects with persistently elevated levels of GCF βG at baseline, 2-week and 3-month recalls had between 7 and 14 times (dependent upon the algorithm used) increased risk ratio for periodontitis progression.
|Reference||Study design||Study groups||Follow-up|
|Akalιn et al. (2005)||Cross-sectional||26 CP, 18 controls||Routine clinical parameters||GCF, biopsy||SOD||Increased in gingiva, no difference in GCF|
|Akalιn et al. (2007)||Cross-sectional||36 CP, 28 healthy||Serum, saliva, GCF||Lipid peroxidation, malondialdehyde, total antioxidant status||LPO and TOS may play an important role in periodontitis|
|Baltacιoğlu et al. (2008)||Cross-sectional||33 CP, 24 healthy||Serum, GCF||Total protein, protein carbonylation||Elevated protein carbonylation may be a sign of oxidative stress in CP|
|Buduneli et al. (2005)||Cross-sectional||20 gingivitis, 20 CP, 20 healthy controls.||GCF, serum||t-PA, u-PA, PAI-1, PAI-2||GCF PAI-2 concentrations were higher in CP, gingivitis than healthy controls. Periodontal disease rather than smoking seems to affect PA system|
|Chapple et al. (2007)||Prospective||18 CP||3 months after SRP||GCF, plasma||TAOC||GCF TAOC was lower in CP, increased after SRP|
|Chen et al. (2009)||Cross-sectional||25 CP, 24 healthy||GCF, serum||PAF||Higher PAF in GCF, serum in CP|
|Dezerega et al. (2010)||15 progression, 18 CP, 10 healthy||GCF||MCP-3||Higher in CP, higher in active sites than inactive sites, may be a diagnostic marker|
|Dutzan et al. (2009)||Cross-sectional||106 moderate to advanced CP||GCF||IFN-γ||Higher in active sites, may be an indicator for progression|
|Emingil et al. (2006b)||Cross-sectional||18 GAgP, 29 CP, 20 gingivitis, 20 healthy controls.||PD, CAL, BOP, PI||GCF||Laminin-5 gamma2-chain||CP had elevated laminin-5 gamma2-chain levels|
|Fitzsimmons et al. (2010)||Cross-sectional, population based||430 periodontitis, 509 healthy controls||GCF||IL-1β, CRP||Higher levels in periodontitis, GCF levels of IL-1β, CRP may indicate higher susceptibility|
|Gapski et al. (2009)||Multi-centre, prospective, controlled trial||CP SDD+surgery|
|12 months||1° outcome: CAL|
2° outcome: PD, BOP, ICTP
|GCF||ICTP||SDD decreased ICTP effectively|
Provided better clinical outcomes
|Garg et al. (2009)||Intervention||20 healthy|
20 CP (also after SRP)
|6–8 weeks after SRP||GI, PD, CAL||GCF||Cathepsin K||Cathepsin K increased in CP, decreased with SRP|
|Golub et al. (2008)||Prospective, controlled||CP, Postmenopausal women SDD: 64|
|2 years||GCF||ICTP, MMP-8, MMP-1, MMP-13, IL-1β||SDD decreased collagenase activity, ICTP|
|Grant et al. (2010)||Intervention||20 healthy, 20 CP||Before and 3 months after SRP||GCF||Reduced glutathione, oxidized glutathione||GSH, GSSG lower in CP|
GSH: GSSG ratio increased after SRP
|Holzhausen et al. (2010)||Cross-sectional, and intervention||40 moderate CP|
40 severe CP
40 healthy controls
|At baseline and also 1 month after SRP||GCF||Trypsin-like activity, TNF-α, IL-1α, IL-6, IL-8, PAR2||PAR2 was higher in CP than controls. IL-1α, -6, -8|
TNF-α, total proteolytic activity were all higher in CP
|Ikezawa-Suzuki et al. (2008)||Intervention||35 CP||At baseline and after SRP||GCF||TNF-α, TNF receptor types 1 and 2||Ratio of TNF R2: R1 increased after treatment, may be related to clinical periodontal data|
|Jepsen et al. (2003)||Cross-sectional||54 teeth with CP, 11 experimental gingivitis||GCF||lysylpyridinoline, hydroxylysylpyridinoline||Increased hydroxylysylpyridinoline, Lysylpyridinoline may be indicators of ongoing destruction|
|Kardeşler et al. (2008)||Cross-sectional||17 DM+periodontal disease, 17 healthy+periodontal disease, 17 healthy controls||GCF||PGE2, IL-1β, t-PA, PAI-2||PGE2, t-PA were higher in DM, and periodontal disease groups than in healthy controls|
|Kurtiş et al. (1999)||24 CP, 24 healthy||PD, PI, GI, BOP, CAL||GCF||IL-6||Higher in CP, no correlation with clinical data|
|Lamster et al. (1994)||GCF, saliva, serum||β-glucuronidase||GCF levels correlated with CAL|
|Lin et al. (2005)||Cross-sectional||14 patients (62 sites divided into 4 groups by PD, BOP).||PD, BOP, CAL||GCF||Oncostatin M, IL-6||Total amounts but not concentrations correlated with disease severity.|
|Mogi & Otogoto 2007)||Cross-sectional||66 CP; 20 mild, 24 moderate, 22 severe|
19 healthy controls
|GCF||Cathepsin K||Increased in periodontitis versus control subjects, (+) correlation between cathepsin K and RANKL suggesting that both contribute to osteoclastic bone destruction|
|Mogi et al. (2004)||Cross-sectional||GCF||Cathepsin K, RANKL||Both increased in CP|
|Nakamura-Minami et al. (2003)||Intervention||21 CP||4 weeks after SRP||Routine clinical parameters||GCF||α-1 protease inhibitor, secretory leukocyte protease inhibitor||Both seem to be related with healing after SRP|
|Oringer et al. (2002)||Intervention||48 CP; SRP+placebo, SRP+minocycline||Before, 1,3,6 months after SRP||GCF||ICTP, IL-1||ICTP, IL-1 correlated with clinical data, may be disease markers|
|Palys et al. 1998||Cross-sectional||7 healthy and 8 gingivitis 21 CP||BOP, PD, CAL, PI||GCF subgingival plaque||ICTP, 40 subgingival taxa||Positive relation between ICTP and putative pathogens|
|Perinetti et al. (2008)||Intervention||16 CP||15, 60 days after SRP||PD, PI, CAL, BOP||GCF||Alkaline phosphatase||Decreased on day 15, increased on day 60 after SRP|
|Pradeep et al. (2009a, b)||Intervention||20 healthy, 20 gingivitis, 20 CP||CP patients also 6–8 weeks after SRP||PD, CAL, GI||GCF||MCP-1||Increased gradually with severity of periodontal disease, decreased after SRP, correlated with clinical periodontal data; may be a biomarker|
|Pradeep et al. (2009a, b)||Intervention||20 healthy, 20 gingivitis, 20 CP||CP patients also 6–8 weeks after SRP||GCF||IL-18, MCP-1||Higher levels in CP, +correlation between IL-18 and MCP-1; IL-18 may induce MCP-1, promote inflammation|
|Rescala et al. (2010)||Cross-sectional||20 GCP, 17 GAgP, 10 gingivitis||GCF||IL-1β, IL-2, IL-4, IL-8, elastase, IFN-γ||IL-1β, elastase were higher in deep sites than shallow sites in periodontitis groups. No significant difference between CP and AgP in GCF levels of biomarkers|
|Rosin et al. (1999)||Cross-sectional||140 subjects||PI, SBI, GCF flow||Saliva, GCF||Peroxidise, lysozyme||No significant correlation between clinical periodontal indices and salivary data|
|Srinath et al. (2010)||Cross-sectional||15 healthy, 15 gingivitis, 15 CP||GI, Russell periodontal index||GCF, saliva||Melatonin||Lowest in CP, may have a protective role in periodontal disease|
|Surna et al. (2009)||Cross-sectional||Healthy|
|PD||GCF, saliva||Lysozyme activity||Lysozyme activity correlated with PD|
|Teles et al. (2010)||Cross-sectional||20 healthy, 20 CP||GCF; subgingival plaque||IL-1β, IL-8, MMP-8, 40 bacterial taxa||+ correlation between clinical indices, GCF cytokines, orange and red complexes|
Healthy sites of CP had higher cytokine levels than healthy group
|Thorat Manojkumar et al. (2010)||Intervention||20 healthy, 20 gingivitis, 20 CP||CP patients also 8 weeks after SRP||PD, CAL||GCF||Oncostatin M||Highest in CP,+correlation with PD, CAL; can be a biomarker in periodontitis|
|Toker et al. (2008)||Intervention||15 healthy, 15 generalized AgP||6 weeks after SRP||PD, CAL, PI, BOP, GI||GCF||IL-1β, IL-1ra, IL-10||IL-1β is high and decreased after SRP|
|Tsalikis et al. (2001)||Intervention||12 advanced periodontitis||4 weeks after SRP||PD, CAL, BOP||GCF||Aspartate amino transferase||AST decreased after SRP|
|Türkoğlu et al. (2009)||Cross-sectional||PD, CAL, PI, BOP, PBI||GCF||IL-18, cathelicidin LL-37||Cathelicidin LL-37 increased in CP, may play a role in destruction|
|Tüter et al. (2007)||Cross-sectional||20 CP, 17 healthy||Serum, GCF||hsCRP||No difference between groups|
|Yetkin Ay et al. (2009)||Cross-sectional||40 CP, 20 healthy||GI, PD, CAL, BOP, PI||GCF||IL-11, IL-17||Healthy group had highest|
IL-11: IL-17 ratio
|Yin et al. (2000)||Intervention||Periodontitis, gingivitis, healthy||14 days after SRP||GCF||t-PA, PAI-2||Decreased after SRP, may be diagnostic markers|
|Yoshinari et al. (2004)||Intervention||7 CP||Before and after SRP||Routine clinical parameters||GCF, gingiva||IL-1α, IL-1β, IL-1ra||IL-1 is effective for evaluating the gingival inflammation|
|Zheng et al. (2006)||Cross-sectional||21 CP, 19 gingivitis, 20 healthy controls||GCF, serum||PAF||PAF levels were higher in CP both in GCF and serum. Serum and GCF levels correlated also with clinical parameters. PAF may play a role in pathogenesis|
Saliva is a mirror of the body that can be used to monitor the systemic as well as the oral health status. Whole saliva contains constituents from exocrine glands secreting into the oral cavity, GCF and dietary and oral plaque components. Saliva is readily available and easily collected without specialized equipment or personnel. Several mediators of chronic inflammation and tissue destruction have been detected in whole saliva of periodontitis patients as described in relatively recent reviews (Kaufman & Lamster 2000, Kinane & Chestnutt 2000, Lamster et al. 2003). In addition, as whole saliva represents a pooled sample with contributions from all periodontal sites, analysis of biomarkers in saliva may provide an overall assessment of disease status as opposed to site-specific GCF analysis (Miller et al. 2006).
Recently, the use of whole saliva as a means of evaluating host-derived products (e.g. salivary gland product, gingival crevice fluid, host enzymes) as well as exogenous components (e.g. oral microorganisms and microbial products) has been suggested as a potential diagnostic marker for disease susceptibility (Şahingür & Cohen 2004). The source of saliva, types of saliva samples, as well as saliva collection methods has been described in detail by Şahingür & Cohen (2004). Development of tests based on the detection of neutrophil defects, genetic markers or the detection and measurement of antibodies specific for periodontal pathogens may be useful in the future. However, there is currently insufficient evidence available for the predictive value of diagnostic tests assessing the host's susceptibility to future periodontitis progression.
Salivary enzymes originate from three major sources; the actual salivary secretions, the host cells found in GCF, and finally disposed bacterial cells from dental plaque and mucosal surfaces. Data from the selected studies evaluating the potential utility of salivary components as diagnostic markers for periodontal tissue destruction are presented in Table 2.
|Reference||Study design||Study groups||Follow-up period||Clinical parameter||Biological sample||Biological parameter||Results|
|Aemaimanan et al. (2009)||Cross-sectional||30 localized CP,|
30 generalized CP,
|PD, CAL, BOP||Saliva, subgingival plaque||Alanine aminopeptidase|
Dipeptidyl peptidase IV, P. gingivalis
|Dipeptidyl peptidase IV but not alanine aminopeptidase activity is associated with CP, presence of P. gingivalis|
|Aurer et al. 1999||Cross-sectional||20 RPP, AP,||Saliva||IL-6||IL-6 increased in RPP, CRP decreased in AP|
|Baron et al. (1999)||Cross-sectional||8 PD patients,|
Total cystatin activity
|Total cystatin activity decreased in patients|
|Cutando et al. (2006)||Cross-sectional||37 patients, variable PD||CPI||Whole saliva||Melatonin||Inverse relation between salivary melatonin and PD severity|
|De la Pena et al. (2007)||Cross-sectional||175 volunteers||Number of teeth, PD||Saliva||Lactate dehydrogenase||Increased in periodontitis, correlated with PD, may be a biomarker|
|Diab-Ladki et al. 2003||Cross-sectional||17 PD, 20 controls||Saliva||TAC, total protein content||TAC decreased, an imbalance between oxidants and antioxidants|
|Garito et al. (1995)||Cross-sectional||69 Periodontal disease patients||PD||Saliva||PAF||Salivary PAF correlated with PD severity|
|Ghallab & Shaker 2010)||Intervention||22 CP, 22 healthy (half smoker, half non-smoker)||At baseline, 1 month after SRP||PI, GI, PD, CAL only in CP||Saliva||sCD44||Highest in smoker CP, decreased after SRP in both smokers and non-smokers|
|Gheren et al. (2008)||Prospective||18 CP, 18 controls||30 days after perio. therapy||PD, CAL, PI, GI||Saliva||Salivary arginase activity||Salivary arginase activity increased in CP, decreased after therapy: a candidate salivary marker of periodontal status|
|Gürsoy et al. (2009)||Cross-sectional||84 CP, 81 controls||Saliva||Elastase, lactate dehydrogenase|
IL-6, IL-1β, TNF-α
|Only IL-1β can differentiate periodontitis|
|Haigh et al. (2010)||Cross-sectional, intervention||Severe periodontitis||Saliva||Proteomics|
|S100 proteins highly related with PD severity|
|Lamster et al. (2003)||Cross-sectional||380 subjects||PD, CAL, GI||Saliva, blood||β glucuronidase||Periodontal clinical parameters correlated with salivary β glucuronidase|
|Koss et al. 2009||Cross-sectional||89 mild, moderate, severe CP|
|Whole saliva||Peroxidase, hydroxyproline, sIgA||Peroxidase, hydroxyproline increased, sIgA diminished in CP|
|Nomura et al. (2006)||Cross-sectional||187 subjects||Saliva||Lactate dehydrogenase, AST, blood urea nitrogen||Salivary lactate dehydrogenase, AST, blood urea nitrogen have potential to screen PD|
|Özmeric et al. (2000)||Cross-sectional||20 CP, 15 controls||PD, CAL, PI, GI||Saliva||Total protein, arginase||Salivary arginase increased in CP|
|Su et al. (2009)||Cross-sectional||58 periodontitis|
|CPITN||Whole saliva||ROS, TAC||ROS increased in periodontitis, TAC correlates with disease severity|
|Takane et al. (2002)||Cross-sectional||78 CP, 17 healthy||Saliva||8-OHdG||Higher in CP, can be a diagnostic marker|
|Teles et al. (2009)||Cross-sectional||74 CP, 44 healthy||Saliva multiplex||G-MCSF, IL-1β, -2, -4, -5, -6, -8, -10|
|No difference between groups. IL-8 correlated with PD, BOP, these cytokines cannot discriminate CP and healthy|
|Totan et al. (2006)||Cross-sectional||Periodontitis, controls||PD, BOP||Saliva||AST, ALP, aminopeptidases, glucuronidases||Salivary AST, ALP increased with pockets, BOP, suppuration|
|Wilczynska-Borawska et al. (2006)||Cross-sectional||26 PD, 20 healthy controls||GI, PBI, PI, PD, CAL, number of teeth||Whole saliva||Hepatocyte growth factor||Salivary HGF may be related with severity of PD|
|Yoshie et al. (2007)||Intervention||49 CP||4 weeks after SRP||PD, CAL, BOP||Saliva||AST, ALT, LDH||All decreased after SRP, they can be useful markers for inflammation, destruction|
Various studies have evaluated the molecular markers of tissue destruction in serum or plasma: these manifestations of periodontal diseases are mainly sought to clarify the possible interactions between periodontitis and various systemic diseases and/conditions such as cardiovascular diseases (CVDs), pregnancy complications, diabetes mellitus, and rheumatoid arthritis. Serum or plasma provides information about the inflammatory stimulus and/or response generated in circulation towards the periodontal pathogens that colonize in the subgingival area (Pussinen et al. 2007). Increased circulating levels of cytokines were reported in chronic periodontitis patients compared with the clinically healthy control subjects (Mooney & Kinane 1994). Data from the selected studies evaluating the potential utility of serum components as diagnostic markers for periodontal tissue destruction are presented in Table 3. Researchers have mainly focused on C-reactive protein (CRP), fibrinogen, and serum amyloid A (SAA) concentrations in serum in relation to periodontitis.
|Reference||Study design||Study groups||Follow-up period||Clinical parameter||Biological|
|Al-Ghamdi & Anιl (2007)||Cross-sectional||30 smoker CP, 30 non-smoker CP, 30 healthy||Serum||Immunoglobulins||Smoking decreases Ig content|
|Amarasena et al. (2008)||Follow-up||266 elderly||6 years Progression: CAL increase >3mm||PD, CAL||Serum||Albumin, Ca, IgG, A, M||Serum Ca correlated with PD, may be a risk factor for progression|
|Behle et al. (2009)||Intervention||30 CP||Before, 4-weeks after SRP||Plasma||19 biomarkers, multiplex||Overall reduction in systemic inflammation, great personal variation|
|De Queiroz et al. (2008)||Cross-sectional||17 CP, 8 healthy||Serum||24 cytokines with multiplex||RANTES differed between groups|
|Duerte et al. (2010)||Intervention||14 healthy controls,|
|At baseline and 6 months after SRP||Serum||IL-17, IL-4, IL-23, IFN-γ, TNF-α||IL-17, TNF-α levels were higher in GAgP than CP, healthy both at baseline and after SRP|
|Furugen et al. (2008)||Cross-sectional||158 Japanese elders||Serum||Adiponectin, resistin, IL-6, TNF-α||Increased serum cortisol is associated with BOP|
|Glas et al. (2008)||Cross-sectional||105 CP, 122 healthy||Plasma||Surfactant protein D||Increased in CP; may be a biomarker for CP|
|Guentsch et al. (2009)||Cross-sectional||425 non-smoker, systemically healthy adults||Number of teeth with PD>4–6 mm||Serum||IL-6, TNF-α||Serum IL-6 is associated with periodontal disease severity|
|Ishisaka et al. (2008)||Cross-sectional||467 adults||PD, CAL, BOP||Serum||Cortisol||Serum cortisol associated with PD, CAL|
|Iwasaki et al. (2008)||Follow-up||600 elderly||4 years||CAL||Serum||Albumin||Serum albumin may be a risk predictor for progression|
|Liu et al. (2010)||Cross-sectional||40 CP, 40 healthy||PD, CAL, BOP||Serum||CRP, lipid profile||CRP higher in CP, HDL lower in CP. correlated with clinical parameters|
|Marcaccini et al. (2009)||Intervention||25 CP, 20 healthy||3 months after SRP||PD, BOP, CAL||Serum, plasma||hsCRP, IL-6, CD40 ligand, MCP-1, sP-selectin, sVCAM-1, sICAM-1||SRP decreased CRP and IL-6, others were not affected|
|Nakajima et al. (2010)||Intervention||78 CP, 40 healthy||At baseline, after SRP||Serum||hsCRP, IL-6, TNF-α||CRP, IL-6 higher and TNF lower in CP, IL-6, CRP decreased after SRP|
|Nibali et al. (2007)||Cross-sectional||302 CP, 183 healthy||Serum||Inflammatory, metabolic markers||CP may be related with systemic inflammation, dysmetabolic state|
|Nicu et al. (2009)||Cross-sectional||105 CP, 57 healthy||Serum||CRP, sCD14||sCD14 increased in CP|
|Offenbacher et al. (2009)||RCT||151 SRP, 152 community care||6 months||PD, CAL, BOP, subgingival calculus||Serum||CRP||Serum CRP was not affected by SRP|
|Pradeep et al. (2010)||Intervention||20 healthy, 20 gingivitis, 20 CP||CP patients also 6–8 weeks after SRP||PD, CAL, GI||Serum||Oncostatin M||Highest in CP, decreased after SRP, not detectable in gingivitis, healthy. Correlated with PD, CAL|
|Raunio et al. (2007)||Cross-sectional||52 subjects||PD, CAL, BL||Mouth-wash, serum||IL-6||Serum IL-6 increased with periodontal disease severity|
|Raunio et al. (2009)||Cross-sectional||56 CP, 28 healthy||Serum||sCD14||Higher in CP, correlated with periodontal disease extent|
|Renvert et al. (2009)||RCT||28 CP test (anti-inflammatory therapy), 29 CP placebo||PD, BOP||Serum||hsCRP, IL-6, IL-1β, IL-8, IL-12, TNF-α, IFN-γ||CRP, IFN, IL-6 decreased, others did not change|
|Trindade et al. (2008)||Cross-sectional||29 CP, 12 AgP, 22 gingivitis, 26 healthy||Serum||Antibody levels against P. gingivalis||Serum Ig titres correlated with clinical periodontal diagnosis|
|Yoshihara et al. (2009)||Cross-sectional||148 subjects at age 77||CAL||Serum||Osteocalcin, bone-specific alkaline phosphatase||Negative correlation between CAL>6 mm and osteocalcin|
The ratio of two molecules has gained attention for being indicative of the regulation of normal bone resorption and deposition activities that occur during bone remodelling, that is the ratio of receptor activator of NF-κB ligand (RANKL) to osteoprotegerin (OPG) (Cochran 2008, Leibbrandt & Penninger 2008). RANKL, receptor activator of NF-κB (RANK), and OPG interactions are important in coordinating osteoclastogenesis and alveolar bone resorption (Suda et al. 1999). RANKL is expressed by osteoblasts/stromal cells (Yasuda et al. 1998), fibroblasts (Quinn et al. 2000), and activated T cells (Horwood et al. 1999). It binds directly to RANK on the surface of preosteoclasts and osteoclasts. RANKL stimulates both the differentiation of osteoclast progenitors and activity of mature osteoclasts (Lacey et al. 1998). This ligand can be found as a cell membrane-bound variant (mRANKL) or in a primary soluble (secreted) form, which has been described in activated T cells (Kong et al. 1999). OPG is the naturally occurring inhibitor of osteoclast differentiation. It is a soluble molecule that binds to RANKL with high affinity and blocks RANKL from interacting with RANK (Lacey et al. 1998). As part of an inflammatory response, pro-inflammatory cytokines, such as interleukin-1β (IL-1β), -6, -11, -17, and tumour necrosis factor-α (TNF-α) can induce osteoclastogenesis by increasing the expression of RANKL while decreasing OPG production in osteoblasts/stromal cells (Nakashima et al. 2000). On the contrary, anti-inflammatory mediators such as IL-13 and interferon-γ (IFN-γ) may lower RANKL expression and/or increase OPG expression to inhibit osteoclastogenesis (Nakashima et al. 2000). Moreover, in a cross-sectional study, Sakellari et al. (2008) reported that soluble RANKL levels in GCF samples of the chronic periodontitis patients were higher than the periodontally healthy subjects, which also correlated with Treponema denticola and Porphyromonas gingivalis levels in the subgingival plaque samples of the same periodontitis patients. However, they failed to show a correlation between GCF RANKL levels and clinical periodontal measurements, namely PD, gingival recession and BOP either cross-sectionally or longitudinally. The concentrations of RANKL and OPG show great variation from study to study, but the ratio of RANKL/OPG has a consistent tendency to increase from periodontal health to periodontitis and to decrease following non-surgical periodontal treatment (Table 4) (Crotti et al. 2003, Mogi et al. 2004, Kawai et al. 2006, Lu et al. 2006, Bostancι et al. 2007, Wara-Aswapati et al. 2007, Buduneli et al. 2008, 2009, Dereka et al. 2010). Thus the ratio of RANKL/OPG shows promise as a discloser of periodontal disease activity.
|Reference||Study design||Study groups||Follow-up|
|Bostancι et al. (2007)||Cross-sectional||21 healthy, 22 gingivitis, 28 CP, 25 GAgP, 11 CP immunosuppressed||GCF||RANKL, OPG||RANKL levels increased, OPG decreased in periodontitis versus gingivitis, healthy. RANKL/OPG ratio may predict disease occurrence|
|Bostancι et al. (2008)||Cross-sectional||21 healthy, 22 gingivitis, 28 CP, 25 GAgP, 11 CP immunosuppressed||GCF||TACE||GCF TACE levels were higher in periodontitis, TACE showed (+) correlation with PD, CAL, GCF RANKL concentration|
|Buduneli et al. (2008)||Cross-sectional||67 untreated CP, 44 maintenance patients||Saliva||sRANKL, OPG||Salivary sRANKL, OPG may be to be affected by smoking and showed significant differences between treated versus untreated CP|
|Buduneli et al. (2009)||Intervention||10 smoker, 10 non-smoker CP||4 weeks after SRP||GCF||IL-17, sRANKL, OPG||OPG decreased, IL-17 increased, sRANKL always similar|
|Frodge et al. (2008)||Cross-sectional||35 moderate to severe CP, 39 healthy controls||Saliva||TNF-α, ICTP, RANKL||TNF-α detected in all subjects, ICTP, RANKL were detected only in a minority. TNF-α was higher in CP than controls. Subjects with salivary TNF-α levels above a threshold had more sites with BOP, PD>4 mm, CAL>2 mm. Salivary TNF-α may be a biomarker for periodontal destruction|
|Golub et al. (2008)||Prospective, controlled||CP, Post-menopausal women SDD: 64|
|2 years||GCF||ICTP, MMP-8, MMP-1, MMP-13, IL-1β||SDD decreased collagenase activity, ICTP|
|Golub et al. (2010)||Prospective||128 postmenopausal women with CP|
SDD or placebo in maintenance
|2 years||Serum||Osteocalcin, ICTP, CTX, bone-specific alkaline phosphatase||SDD decreased significantly serum biomarkers of bone resorption|
|Gürlek et al. (2009)||Cross-sectional||34 smokers, 22 non-smokers, 11 ex-smokers||Saliva||ICTP, OC, cotinine||ICTP was similar in study groups, OC was lower in smoker than the other groups, correlated negatively with pack-years|
|Mogi et al. (2004)||Cross-sectional||GCF||Cathepsin K, RANKL||Both increased in CP|
|Oringer et al. (2002)||Intervention||48 CP; SRP+placebo, SRP+minocycline||Before, 1,3,6 months after SRP||GCF||ICTP, IL-1||ICTP, IL-1 correlated with clinical data, may be disease markers|
|Özçaka et al. (2010)||Cross-sectional||58 CP, 47 healthy controls||plasma||ICTP, osteocalcin||CP and healthy control groups were similar in plasma ICTP, OC levels|
|Palys et al. (1998)||Cross-sectional||7 healthy, 8 gingivitis|
|BOP, PD, CAL, PI||GCF subgingival plaque||ICTP, 40 subgingival taxa||Positive relation between ICTP and putative pathogens|
|Sakellari et al. (2008)||Cross-sectional||35 CP, 38 periodontally healthy||PD, recession, BOP||GCF, sub-gingival plaque||Free sRANKL, Aggregatibacter actinomycetemcomitans, Tannerella forsythensis Treponema denticola||sRANKL was higher in CP, correlated with Treponema denticola, P. gingivalis but not with clinical parameters|
|Silva et al. (2008)||Longitudinal||56 CP moderate to severe, followed until progression||GCF||RANKL, MCP-1, TNF-α, IL-1β, MMP-13||Higher RANKL, IL-1β, MMP-13 activity in active sites|
Alkaline phosphatase (ALP) is a catalysing enzyme that accelerates the removal of phosphate groups in various molecules. ALP was not supported to have a predictive value for periodontal breakdown, but it may serve as a marker in periodontal treatment planning and monitoring (McCauley & Nohutçu 2002, Taba et al. 2005). Aspartate amino transferase (AST) was found to be increased in GCF of severe periodontitis patients decreasing after initial periodontal treatment (Tsalikis et al. 2001).
GCF oncostatin M levels were highest in chronic periodontitis patients when compared with gingivitis patients and healthy control subjects (Thorat Manojkumar et al. 2010). Serum oncostatin M levels were also associated with PD and CAL in chronic periodontitis patients (Pradeep et al. 2010). The positive correlations between oncostatin M levels and PD as well as CAL suggest that it may be a useful biomarker in periodontitis.
Cathepsin B is a proteinase which is synthesized mainly by macrophages in GCF and levels of cathepsin B may distinguish gingivitis from periodontitis (Kennett et al. 1997). Furthermore, GCF levels of cathepsin B correlate significantly with clinical parameters before and also after periodontal treatment, suggesting a use for this enzyme in treatment planning and monitoring (Chen et al. 1998).
Cathepsin K, a cysteine protease, is capable of hydrolysing extracellular bone matrix proteins, is highly expressed in osteoclasts, and it is a well-known marker of osteoclast activity (Motyckova et al. 2001). Eley & Cox (1996a, b) have suggested that GCF cathepsin B level may indeed serve as a predictor of attachment loss. Mogi & Otogoto (2007) investigated GCF levels of cathepsin K and RANKL in 20 mild, 24 moderate, and 22 severe chronic periodontitis patients in comparison with 19 healthy control subjects. Cathepsin K was below the detection limit in the healthy control group, whereas it was detectable in all periodontitis samples. In a recent intervention study, Garg et al. (2009) reported that GCF levels of cathepsin K was higher in chronic periodontitis patients than gingivitis patients and healthy controls, whereas it was decreased significantly after scaling and root-planing. The high levels of cathepsin K in periodontitis samples together with the positive correlation between cathepsin K and RANKL levels suggest that both of them contribute to osteoclastic bone resorption in periodontitis, although no significant correlation could be found between the biomarkers and clinical periodontal measurements.
Carboxyterminal telopeptide pyridinoline cross-links of type I collagen (ICTP) is released into the periodontal tissues as a consequence of collagen degradation and alveolar bone resorption (Seibel 2003). Type I collagen composes 90% of the organic matrix of bone and is the most abundant collagen in osseous tissue (Narayanan & Page 1983). Studies assessing the role of ICTP levels in GCF or peri-implant crevicular fluid as a diagnostic marker of periodontal disease activity have had promising results so far (Oringer et al. 1998, 2002). ICTP was suggested to predict future bone loss, to correlate with clinical parameters and putative periodontal pathogens and also to reduce following periodontal therapy, thus leading to an accurate assessment of tissue breakdown (Giannobile 1999). Because of its specificity and sensitivity for bone resorption, ICTP represents a potentially valuable diagnostic marker for periodontal disease (Giannobile et al. 2003).
Osteocalcin (OC) is a calcium-binding protein of bone and the most abundant non-collagenous protein of the mineralized tissue (Lian & Gundberg 1988). Osteoblasts, odontoblasts, and chondrocytes produce OC (Gallop et al. 1980). Serum level of OC is considered as a marker of bone formation (Christenson 1997, Giannobile et al. 2003). Lower serum levels of OC were reported in periodontitis patients compared with healthy subjects suggesting lower osteoblastic activity and bone formation ability (Shi et al. 1996). Yoshihara et al. (2009) reported that serum OC levels correlated with CAL in elderly Japanese subjects. On the other hand, indifferent plasma levels of ICTP and OC were reported in a recent study comparatively evaluating chronic periodontitis patients and healthy control subjects (Özçaka et al. 2010). There is no consensus, yet on the potential usage of OC for assessment of periodontal diagnosis and prognosis.
Osteonectin is a single-chain polypeptide that binds strongly to hydroxyapatite and other extracellular matrix molecules like collagens. It may be a sensitive marker for detection of periodontal disease status. However, no clinical study reporting osteonectin levels in biological fluids in relation with periodontitis could be found in the present literature search.
Osteopontin (OPN), a glycosylated phosphoprotein, is a bone matrix component produced by osteoblasts, osteoclasts, and macrophages as a multi-functional cytokine. It is a single-chain polypeptide which is concentrated at clear zone attachment areas of plasma membrane where osteoclasts are attached to the underlying mineral surface that is, the clear zone attachment areas of the plasma membrane (Rodan 1995). Kido et al. (2001) showed the presence of OPN in GCF and also showed that GCF OPN levels increased parallel to the increase in PD. Sharma & Pradeep (2006) reported that GCF OPN concentrations increased proportionally with the progression of disease and decreased significantly following non-surgical periodontal treatment. The same authors reported that chronic periodontitis patients exhibited the highest GCF and plasma OPN levels which decreased significantly following initial periodontal treatment (Sharma & Pradeep 2007). The GCF OPN levels correlated with plasma levels and clinical attachment loss suggesting a role in the pathogenesis of periodontitis. Thus, OPN seems to be a promising biomarker for periodontal disease progression.
Matrix metalloproteinases (MMPs) are considered as modifiers of host response and it has been suggested that their role and involvement should not be interpreted solely as surrogate promoters of tissue destruction, but also as defensive or protective factors against inflammation as a whole (Sorsa et al. 2004, 2006). MMPs represent a structurally related but genetically distinct superfamily of proteases acting not only in physiological development and tissue remodelling but also in pathological tissue destruction (Sorsa et al. 2004). MMPs can be divided into five major groups: collagenases (MMP-1, -8, -13), gelatinases (MMP-2, MMP-9), stromelysins (MMP-3, -10, -11), membrane-type MMPs (MMP-14, -15, -16, -17), and others (Sorsa et al. 2004). MMPs can collectively degrade almost all components of extracellular matrix and basement membrane and their excess activity lead to periodontal tissue destruction. The interplay of cell–cell and cell–matrix interactions involving the production of enzymes, activators, inhibitors, cytokines, and growth factors regulate not only connective tissue remodelling but also connective tissue matrix destruction (Reynolds & Meikle 1997). MMPs produced by resident cells including fibroblasts, macrophages, neutrophils, and epithelial cells, are controlled by tissue inhibitors of MMPs (TIMPs). MMPs can also process bioactive non-matrix substrates such as cytokines, chemokines, growth factors, and immune modulators thereby mediating anti-inflammatory and pro-inflammatory processes (Sorsa et al. 2006, Kuula et al. 2009). Upon bacterial insult, triggered leucocytes migrate to the site of inflammation and release MMP-8 and MMP-9, which are activated locally (Sorsa et al. 2006). TIMPs regulate the activities of these enzymes and TIMP-1 is more effective on interstitial collagenases. An imbalance between MMPs and TIMPs result in the pathological tissue destruction observed in periodontitis (Table 5) (Aiba et al. 1996, Bιyιkoğlu et al. 2009, Kardeşler et al. 2010).
|Reference||Study design||Study groups||Follow-up|
|Alfant et al. (2008)||Cross-sectional||44 children with/without AgP|
|PD||GCF||MMP-1, -2, -3, -8, -9, -12, -13||MMP levels higher in AgP than the others, especially in diseased sites|
No correlation with PD
|Alpagot et al. (2001)||Prospective||40 CP||6 months||GI, PI; BOP, PD, CAL, suppuration||GCF||MMP-3, TIMP-1||High GCF MMP-3, TIMP-1 may indicate high risk for progression|
|Beklen et al. (2006)||Cross-sectional||CP||GCF||MMP-3, -8, -9||GCF MMP-8, -9 levels correlated with disease activity|
|Bildt et al. (2008)||Cross-sectional||8 healthy, 12 CP||GCF||MMPs, TIMPs||MMPs higher, TIMPs lower in CP, with high individual variation|
|Emingil et al. (2006c)||Cross-sectional||35 GAgP, 29 CP, 20 gingivitis, 21 healthy controls.||PD, CAL, BOP, PI||GCF||MMP-25, MMP-26||MMP-25, -26 levels increased in both periodontitis groups and associated with severity of periodontal inflammation, suggesting that they can have a role in disease progression|
|Emingil et al. (2006a)||Cross-sectional||20 GAgP, 20 CP, 20 gingivitis, 20 healthy controls||PD, CAL, BOP, PI||GCF||MMP-7, TIMP-1, EMMPRIN||MMP-7 levels were similar in groups. All patient groups had higher EMMPRIN, TIMP-1. Periodontitis groups had higher TIMP-1 than gingivitis. Increased EMMPRIN, TIMP-1 related with severity of periodontal inflammation|
|Gürsoy et al. (2010)||Cross-sectional||84 CP, 81 controls||Saliva||MMP-8, -14|
|MMP-8, TIMP-1, ICTP increased in CP and can differentiate periodontitis|
|Heikkinen et al. (2010)||Cross-sectional||258 boys, 243 girls all adolescents, smoker versus non-smokers||Saliva||MMP-8, elastase||Median MMP-8 and elastase were lower in smokers, boys seem to be more susceptible than girls|
Both biomarkers correlated with BOP
|Hernandez et al. (2006)||Prospective||21 CP|
Active and inactive sites
|GCF||MMP-13||Active sites had higher MMP-13; a marker of disease progression|
|Hernandez-Rios et al. (2009)||Prospective, active sites defined in 2 months||Moderate-advanced CP||PD, CAL, BOP||GCF||MMP-9, MMP-13, TIMP-1||MMP-13 activity significantly elevated in active sites, high levels of ICTP, associated with progression of periodontal breakdown. MMP-9, -13 may be useful biomarkers for periodontitis progression|
|Kardeşler et al. (2010)||Cross-sectional||12 DM gingivitis, 12 DM periodontitis, 12 H gingivitis, 13 H periodontitis, 24 healthy controls.||GCF||MMP-8, MMP-13, TIMP-1||MMP-8 total amounts were higher in periodontitis, DM-G groups. MMP-13, TIMP-1 total amounts were similar in study groups|
|Kinane et al. (2003)||Intervention||20 CP, before and after SRP||3 months||PD, CAL||GCF||MMP-8||MMP-8 decreased with SRP|
|Mäntylä et al. (2003)||Intervention||11 CP, 10 gingivitis, 8 healthy controls||Before and after SRP||Routine clinical parameters||GCF||MMP-8||Chair-side MMP-8 can differentiate CP from gingivitis, healthy, and monitor treatment of CP|
|Mäntylä et al. (2006)||Prospective||CP|
Smoker versus NS
|12 months||GCF||MMP-8||Persistent elevation in MMP-8 may indicate high risk, poor response to therapy|
|Marcaccini et al. (2009)||Intervention||28 CP,|
|before and 3 months after SRP||Plasma||MMP-2, -3, -8, -9, TIMP-1, -2, total gelatinolytic activity||Higher MMP-3, -8, -9, gelatinolytic activity in CP. MMP-8, -9, gelatinolytic activity decreased after therapy|
|Miller et al. (2006)||Cross-sectional||28 moderate-severe CP, 29 healthy controls||Saliva||IL-1β, MMP-8, OPG||Elevated IL-1β, MMP-8 but not OPG significantly increased risk of periodontal disease. Their salivary levels may be biomarkers for periodontal disease|
|Passoja et al. (2008)||Cross-sectional||48 CP patients||PI, PD, CAL, BOP||GCF, serum||MMP-8||MMP-8 in shallow crevices associated with CAL, can be a prognostic marker|
|Pozo et al. (2005)||Intervention||13CP, 11 healthy||6 months, after SRP||Routine clinical parameters||GCF||MMP-8, -9, TIMP-1, -2||Significant+correlations between periodontal disease severity and MMP activity, – correlations with TIMP-1, -2|
|Rai et al. (2008)||Cross-sectional||15 healthy|
|GCF, saliva||MMP-2, -8, -9||Salivary MMP-8, GCF MMP-9 were higher in CP|
GCF MMP-2 was lower in CP
High correlation with PD, BOP
|Ramseier et al. (2009)||Cross-sectional||50 healthy/gingivitis|
|Whole saliva||MMP-8, -9, OPG||MMP-8, -9, OPG combined with red complex predicted periodontal disease|
|Xu et al. (2008)||Cross-sectional||Healthy, gingivitis, CP||GCF||MMP-8 isoenzymes (fibroblast and PMN-type)||CP exhibited highest activation of GCF MMP-8 isoenzymes|
Salivary TIMP-1 concentration has been reported to be lower in periodontitis patients than the healthy controls, while collagenase activity was higher, which exhibited reciprocal changes after initial periodontal therapy (Hayakawa et al. 1994). Furthermore, saliva samples from healthy subjects consisted mainly procollagenase whereas active collagenase predominated in diseased subjects. Greater collagenase activity has been found also in GCF samples of periodontitis patients than the healthy subjects (Villela et al. 1987). In a prospective study on chronic periodontitis patients comparing active versus inactive sites, significantly higher GCF levels of MMP-13 were reported in active sites, suggesting that it could serve as a marker of disease progression (Hernandez et al. 2006).
Ramseier et al. (2009) reported from a recent cross-sectional study that salivary concentrations of MMP-8, -9, and OPG combined with red complex bacteria predicted periodontal disease. In another recent study (Hernandez-Rios et al. 2009), active sites in moderate to advanced chronic periodontitis patients followed for 2 months were defined and GCF levels of MMP-9 and -13 were suggested as useful biomarkers for periodontitis progression. Another cross-sectional study reported that GCF MMP-8, -9 levels correlated with disease activity in chronic periodontitis patients (Beklen et al. 2006). Supporting data came from an intervention study comprising 28 chronic periodontitis patients and 22 controls (Marcaccini et al. 2009). The authors reported higher MMP-3, -8, -9, and gelatinolytic activities in plasma of chronic periodontitis patients which decreased significantly 3 months after non-surgical periodontal treatment. Similar data on GCF MMP-8 levels were also reported by Chen et al. (2000). Kinane et al. (2003) also reported that GCF MMP-8 levels decreased significantly 3 months after non-surgical periodontal therapy in 20 chronic periodontitis patients. Various treatment modalities have been investigated to determine their potential to control the activities of MMPs in periodontal tissue destruction (Buduneli et al. 2002, 2007, Golub et al. 2008, Vardar-Şengül et al. 2008).
GCF MMP-8 levels in shallow crevices were reported to be associated with CAL and thus it was suggested as a prognostic marker (Passoja et al. 2008). Higher MMP-8 levels in saliva and higher MMP-9 levels in GCF of chronic periodontitis patients were detected in comparison with gingivitis patients and healthy control subjects (Rai et al. 2008). Significant positive correlations were found between periodontal disease severity and GCF MMP-8, -9 activities together with negative correlations with TIMP-1, -2 levels (Pozo et al. 2005). Accordingly, GCF MMP-8 activation was found to be the highest in chronic periodontitis patients compared with healthy subjects and gingivitis patients (Xu et al. 2008). In cross-sectional studies, elevated GCF levels of MMP-7, TIMP-1 (Emingil et al. 2006c), laminin-5 gamma2-chain (Emingil et al. 2006a), MMP-25, -26 (Emingil et al. 2006b) were reported in chronic periodontitis patients compared to gingivitis patients and healthy control subjects. High GCF MMP-13, TIMP-1 levels were suggested to indicate high risk for periodontal disease progression (Alpagot et al. 2001). Similarly, persistent elevation of MMP-8 in GCF samples was regarded as indicating high risk and poor response to periodontal therapy (Mäntylä et al. 2006). Very recently, Gürsoy et al. (2010) also confirmed that salivary MMP-8 and TIMP-1 are increased in chronic periodontitis patients differentiating periodontitis patients from the controls. Furthermore, a chair-side MMP-8 test was indicated to effectively differentiate chronic periodontitis from gingivitis and clinically healthy sites and thus, to monitor treatment of chronic periodontitis patients (Mäntylä et al. 2003).
Neutrophil elastase (NE) is one of the most destructive enzymes with the capability of degrading almost all extracellular matrix components as well as plasma proteins and activating pro-MMPs and inactivating TIMP-1 (Meyle et al. 1992, Eley & Cox 1996a, b, Sorsa et al. 2006, Geraghty et al. 2007). A high concentration of NE is stored in azurophilic granules of PMNs, providing an important step in host defence. When activated, NE can be released rapidly into the extracellular space and cause local tissue damage (Kawabata et al. 2002). Endogenous proteinase inhibitors are important to protect tissues from unregulated proteolysis. Once released in circulation, NE is rapidly inactivated by conjugating with protease inhibitors. Heikkinen et al. (2010) reported that salivary concentrations of MMP-8 and NE correlated with BOP in a cross-sectional study comprising 501 adolescents.
Azurophilic granules of PMNs also contain the enzyme myeloperoxidase (MPO), which can generate a reactive oxidant species including hypochlorous acid (HOCl). MPO is produced by neutrophils whose oxidant products are capable of modifying low-density lipoprotein cholesterol and has been shown to be present in human atheromas and unstable plaques (Nicholls & Hazen 2005). MPO is released into the extracellular environment following neutrophil stimulation and/or degranulation (Buchmann et al. 2002). MPO can oxidatively activate MMP-8 and -9 and inactivate TIMP-1 (Sorsa et al. 2006). Decreased plasma MPO levels following periodontal treatment of severe periodontitis patients suggests that MPO may play a major role in pathogenesis of destructive periodontal diseases (Behle et al. 2009). Thus, MPO and NE can potentiate the destructive MMP cascades and indeed MPO has been regarded as a promising marker of periodontal disease activity (Arbes et al. 1999, Yamalιk et al. 2000, Wei et al. 2004).
Eley & Cox (1995) have investigated whether GCF levels of dipeptidyl peptidase (DPP) II or IV levels, total activity and concentration could predict progressive attachment loss and suggested that both GCF DPP II and IV may be predictors of periodontal attachment loss. In another early study, Cox & Eley (1992) have also analysed cathepsin B/L-, elastase-, tryptase-, trypsin-, and DPP IV-like activities in GCF samples obtained from chronic periodontitis patients before and also after non-surgical periodontal treatment. GCF protease levels were suggested to reflect the clinical status of periodontal lesions and may thus be of value in monitoring disease activity.
Prostaglandin E2 (PGE2), a metabolite of the cyclooxygenase (COX) pathway, is an arachidonic acid metabolite and considered a potent mediator of alveolar bone loss in periodontitis (Offenbacher et al. 1984, 1993). PGE2 is known to have an activity on fibroblasts and osteoclasts to induce the synthesis of MMPs, IL-1β and other cytokines. PGE2 has been detected in higher levels in gingival tissue and GCF proportional to the severity of periodontal disease. IL-1β is a central mediator of inflammation and connective tissue destruction in rheumatoid arthritis (Raymond et al. 2006). IL-1β increases matrix degradation also by inducing the production of PGE2 in synovial cells, as well by its role as a mediator of bone and cartilage destruction (Cutulo 2004). TNF-α is a less potent stimulator of PGE2 production compared to IL-1β, but these two cytokines synergistically enhance PGE2 production (Yücel-Lindberg et al. 1999). Several of the cytokines stimulating RANKL expression and bone resorption also enhance the expression of COX-2 and prostaglandin production (Lerner 2006). PGE2 is also an efficient stimulator of RANKL expression in osteoblasts (Li et al. 2002) as well as on osteoclast progenitor cells (Ono et al. 2005). At persistent high concentrations, PGE2 stimulates the master osteoclast activator; RANKL expression in stromal cells/osteoblasts and eventually causes enhanced bone resorption. Offenbacher et al. (1986) demonstrated that PGE2 levels in GCF of patients exhibiting periodontal diseases are significantly higher than those in periodontally healthy subjects, and furthermore, that GCF PGE2 concentrations are effective for predicting periodontitis progression, i.e. attachment loss, with a high degree of sensitivity and specificity (0.76 and 0.96, respectively). Evidence indicating the role of prostaglandins in periodontal tissue destruction also comes from human studies evaluating potential effects of non-steroidal anti-inflammatory drugs in periodontal treatment (Buduneli et al. 2002, 2010, Vardar et al. 2003).
Pro-inflammatory cytokines also play a significant role in the pathogenesis of periodontal diseases in terms of both soft and hard tissue destruction. Besides IL-1β and TNF-α, IL-6 is also hypothesized to stimulate bone resorption (Ishimi et al. 1990). TNF-α, IL-1β, -6, -8, -18 have been found to be abundantly expressed in human gingiva, and increased levels have been reported in GCF of periodontitis patients (Preiss & Meyle 1994, Boch et al. 2001, Graves & Cochran 2003, Toker et al. 2008, Pradeep et al. 2009a, b, Fitzsimmons et al. 2010, Teles et al. 2010). Increased GCF levels of IL-1β and NE have been reported with increasing level of gingival inflammation also in experimental gingivitis studies and indeed elastase activity has been suggested as an excellent quantitative measure of gingival inflammation (Hermann et al. 2001, Gonzáles et al. 2001). On the contrary, mean salivary levels of granulocyte-macrophage colony-stimulating factor, IL-1β, -2, -4, -5, -6, -8, -10, interferon gamma (IFN-γ), and TNF-α failed to discriminate between periodontal health and disease (Teles et al. 2009). Neutralization of IL-1 and TNF-α by soluble receptors has been reported to decrease osteoclast formation and bone loss in experimental periodontitis (Assuma et al. 1998). Silva et al. (2008) followed 56 moderate to severe chronic periodontitis patients until progression of periodontal destruction was detected and they reported significantly higher GCF levels of RANKL, IL-1β, and MMP-13 in active sites than the inactive sites. Very recently, Rescala et al. (2010) conducted a cross-sectional study comprising 20 chronic periodontitis patients, 17 generalized aggressive periodontitis patients and 10 gingivitis patients. They reported that IL-1β and elastase levels in the GCF samples were higher in deep sites compared with the shallow sites in both of the periodontitis groups and suggested that these biomarkers may indicate periodontal tissue destruction. Furthermore, Frodge et al. (2008) evaluated salivary concentrations of TNF-α, RANKL, and ICTP in 35 subjects with moderate to severe chronic periodontitis in comparison with 39 healthy controls. The authors reported that salivary TNF-α levels were significantly elevated in chronic periodontitis patients suggesting the utility of this biomarker in a panel of salivary parameters that could facilitate the screening, diagnosis, and management of periodontal disease. In another recent study, Yetkin Ay et al. (2009) investigated GCF IL-11 and IL-17 levels in 40 chronic periodontitis patients and 20 healthy controls. It was reported that the IL-11/IL-17 ratio was significantly higher in the healthy control group than the periodontitis patients, whereas shallow sites in the periodontitis patients exhibited higher ratios than the deep sites in the same patients. Thus, an imbalance in the pro- and anti-inflammatory cytokines may be responsible for periodontal breakdown through immune responses.
CRP is a circulating acute-phase protein acting in an innate immune response. Significantly higher serum high-sensitivity CRP (hs-CRP) concentrations were found in periodontitis patients than the healthy controls in various case–control studies (Ebersole et al. 1997, Fredriksson et al. 1999, Loos et al. 2000, Glurich et al. 2002, Craig et al. 2003, Liu et al. 2010). On the other hand, Tüter et al. (2007) found no difference in serum CRP levels between the chronic periodontitis and healthy control groups, and serum CRP levels were not associated with PD. Moreover, periodontal treatment modalities resulted in decreases in serum CRP concentrations (Ebersole et al. 1997, Nakashima et al. 2000, Mattila et al. 2002, Swoboda et al. 2008, Marcaccini et al. 2009, Katagiri et al. 2009, Renvert et al. 2009, Golub et al. 2010). However, Offenbacher et al. (2009) reported that serum CRP concentration was not affected by scaling and root planning.
Serum amyloid A (SAA) is another acute-phase protein, which is associated with high-density lipoprotein (HDL). Elevated SAA concentrations in blood circulation have been found in subjects with both periodontitis and CVDs compared with controls without either disease (Glurich et al. 2002). A positive correlation between SAA level in serum and number of purulent periodontal pockets has been reported (Pussinen et al. 2004). Vuletic et al. (2009) reported that serum SAA concentration decreased after full-mouth extraction in periodontitis patients.
Increased serum levels of IL-1β, TNF, OC, soluble intercellular adhesion molecule (sICAM), IL-6, MMP-9 have been reported in experimental studies in animal models as well as in clinical studies in humans (Raunio et al. 2007, Guentsch et al. 2009, Nakajima et al. 2010). However, Özçaka et al. (2010) failed to find significant differences between chronic periodontitis patients and healthy control subjects in terms of plasma OC and ICTP concentrations. In an intervention study, Marcaccini et al. (2009) reported that chronic periodontitis patients had higher plasma concentrations of MMP-3, -8, -9, and gelatinolytic activity, whereas non-surgical periodontal treatment decreased their levels significantly.
Serum cortisol concentration was reported to be associated with clinical periodontal parameters; PD and CAL (Ishisaka et al. 2008) and with BOP (Furugen et al. 2008). Furthermore, surfactant protein D was suggested to be a biomarker for chronic periodontitis, as it was found to be increased in 105 chronic periodontitis patients in comparison with 122 healthy control subjects (Glas et al. 2008). In a follow-up study, Amarasena et al. (2008) suggested that serum calcium level may be a risk factor for periodontal disease progression, as it correlated with PD. Higher platelet activating factor (PAF) levels were found in both GCF and serum of chronic periodontitis patients than gingivitis patients and healthy control subjects (Zheng et al. 2006, Chen et al. 2009). These elevated levels together with the significant correlations with clinical periodontal measurements suggested that PAF may have a role in the pathogenesis of periodontal destruction.
There exists extensive evidence that molecules in the saliva and in the GCF correlate with tissue inflammation and bone destruction. Among those showing most promise is the ratio of RANKL/OPG, which is a discloser of periodontal disease activity but its utility in predicting future disease is questionable. However, the sensitivity and specificity of these molecules as predictors of future disease, i.e. prognosis, has not been reliably demonstrated. Even after the advent of highly sophisticated methods and almost 30 years of research on finding a consistent and definitive biochemical marker for periodontal disease activity and prognosis, no adequate marker has been successfully identified as yet. The natural history and nature of periodontal disease is one highly complicating factor in this situation as periodontal disease progresses episodically and it is very difficult to define the quiescent and active periods. Moreover, periodontal disease is a multi-factorial disease with several host- and microorganism-related factors acting consecutively. Last but not least, there is a wide range of individual variability in the response to the microbial triggers as well as to the therapeutic measures. Another issue making it difficult to come to a conclusion on the biochemical parameters is the diversity in biological sample collection, storage, and analysis methods used in different studies evaluating the same parameters. We should not be surprized that finding markers for periodontal disease is so elusive: other chronic conditions such as CVDs are similarly complex and despite even greater expenditure and manpower being applied, recent reports reveal that the previously lauded vascular risk factors such as CRP and N-BNP, are in fact not useful (Melander et al. 2009). Interestingly CRP and other risk factors in CVD are more useful as negative predictive factors similar to the use of BOP as a negative predictive factor in periodontal disease (Lang et al. 1986). At present, we do not have a clear understanding of the pathology or the molecular events occurring in the periodontal microenvironment during the tissue destruction process. Certain clinical measures such as clinical attachment loss remain the strongest predictor of future attachment loss and absence of certain clinical inflammatory signs such as BOP are excellent negative predictors of periodontal inflammation. Thus we can conclude that much work remains to identify molecules with clinical utility for estimating current and future destructive periodontal disease activity. We provide the following supportable conclusions, many of which are not novel.
- •Clinical measures such as PD, attachment level and BOP are essential for the diagnosis of prevalent periodontal tissue destruction.
- •BOP remains to be the most reliable clinical finding correlated with periodontal disease activity.
- •An increase in PD over time is associated with loss of attachment and loss of tooth supporting bone. Current PD is indicative of future attachment loss (Van der Velden et al. 2006). Thus, clinical periodontal measurements, mainly PD and BOP continue to be the most reliable parameters not only for the diagnosis but also for determining the prognosis of periodontal diseases, although they do not provide 100% accuracy.
- •Investigating the levels of RANKL, OPG, and a range of host bone destruction markers and inflammation related molecules in GCF and saliva may prove to be reliable information on the state of periodontal disease activity however, currently cannot predict future disease activity.
- •With the current picture, it is quite clear that highly specific and sensitive biomarkers for diagnosis and monitoring of periodontal diseases are still needed for early and better detection of periodontal tissue destruction.
Scientific rationale for the study: Objective and ideal diagnostic methods for periodontal diagnosis of the present as well as future tissue destruction are still being sought.
Principal findings: Clinical periodontal measurements, mainly PD and BOP, continue to be the most reliable parameters not only for the diagnosis but also for determining the prognosis of periodontal diseases. A range of host bone destruction markers and inflammation related molecules in GCF, saliva and blood are not reliable to predict future tissue destruction or disease activity.
Practical implications: It is quite clear that highly specific, sensitive biomarkers for monitoring of periodontal diseases are still needed for better detection of periodontal tissue destruction.