Periodontitis is a chronic inflammatory condition of the tissues that surround and support the teeth and is initiated by inappropriate and excessive immune responses to bacteria in subgingival dental plaque leading to loss of the integrity of the periodontium, compromised tooth function, and eventually tooth loss. Periodontitis is an economically important disease as it is time-consuming and expensive to treat. Periodontitis has a worldwide prevalence of 5–15% and the prevalence of severe disease in western populations has increased in recent decades. Furthermore, periodontitis is more common in smokers, in obesity, in people with diabetes, and in heart disease patients although the pathogenic processes underpinning these links are, as yet, poorly understood. Diagnosis and monitoring of periodontitis rely on traditional clinical examinations which are inadequate to predict patient susceptibility, disease activity, and response to treatment. Studies of the immunopathogenesis of periodontitis and analysis of mediators in saliva have allowed the identification of many potentially useful biomarkers. Convenient measurement of these biomarkers using chairside analytical devices could form the basis for diagnostic tests which will aid the clinician and the patient in periodontitis management; this review will summarise this field and will identify the experimental, technical, and clinical issues that remain to be addressed before such tests can be implemented.
Periodontitis and gingivitis are the most common forms of periodontal disease; these disorders are caused by disruption to normal homeostatic processes by numerous bacterial species found in subgingival dental plaque [
There is a global variation in the prevalence, severity, and progression of periodontitis [
Clinical and epidemiological evidence reveals an association between chronic periodontitis and a number of systemic conditions, most notably diabetes and cardiovascular disease (CVD) [
Currently, there are no dependable tests to diagnose and predict progression of periodontitis. Although clinical diagnosis of periodontitis based on visual and radiographic assessment in addition to measurement of pocket depth, tissue attachment, and “bleeding on probing” (BOP) in different locations in multiple teeth is well established in dental practice, these are time-consuming procedures for the periodontist [
The concept of a biomarker arose from the recognition of the appeal of being able to monitor health status, disease susceptibility, progression, resolution, and treatment outcome with respect to a number of common medical conditions [
The holy grail of biomarker research in periodontology is to develop a “high impact diagnostics” which have a significant impact on clinical decision making, patient outcomes, and healthcare economics [
Saliva comprises the components of exocrine secretions of the oral salivary glands as well as GCF and elements of dental plaque and the diet; the anatomy and physiology of saliva secretion and the detailed composition of saliva are reviewed elsewhere [
Salivary diagnostic tests for periodontal disease include those based on analysis of molecular components, periodontal pathogens, and DNA for putative genetic risk [
Although the microbial components of the subgingival plaque biofilm are the aetiological factors, the pathogenesis of periodontitis is a complex interaction between the microbiota and host tissues modified by environmental factors in particular smoking, age, systemic disease, and genetic susceptibility. Our knowledge of the complexity, dynamics, and properties of the dental plaque biofilm is substantial [
There are many molecular signals which regulate inflammatory processes such as periodontitis [
Schematic to illustrate the pathogenesis of periodontitis. The dental plaque biofilm is complex, dynamic, and variable; it is subject to quantitative and qualitative ecological shifts in response to changes in the local environment (e.g., pH changes), changes in localised immune regulation, and extrinsic factors such as smoking. Bacteria in dental plaque signal the local tissue cells and immune cells via intrinsic and secreted microbe-associated molecular patterns (MAMPs) such as lipopolysaccharide (LPS) and specific antigens (e.g., fimbrial proteins). The healthy periodontium is maintained by an effective innate response to a commensal (nonpathogenic) microflora in dental plaque which is restricted to the gingival/plaque margin and in which neutrophils play a pivotal role regulated by low levels of cytokines such as IL-1
Epidemiological studies established that, in a homogenous population with even plaque distribution and oral hygiene practices, periodontal disease experience is not evenly distributed [
The biochemical pathways leading to cytokine secretion during the development of periodontitis involve signalling via the interaction of microbe-associated molecular patterns (MAMPS, e.g., LPS, DNA, fimbriae, etc.) and pattern recognition receptors (PRR, e.g., Toll-like receptors (TLR) and NOD-like receptors (NLR)) (Figure
There is substantial evidence for a key role of numerous cytokines in the initiation and regulation of immune responses in periodontitis, and, critically, in mediating tissue destruction (mainly through activation of host MMPs) [
IL-1
IL-6 is secreted by a wide range of cells in the periodontium, probably as a secondary response to IL-1
Alveolar bone loss is a critical feature of disease progression in periodontitis and there is an increasing recognition of the important regulatory interactions between bone metabolism and inflammation [
There are some 23 MMPs in man which comprise a structurally related but genetically distinct superfamily of zinc-dependent endoproteases [
Collectively, we have much information on the role of a wide range of other host regulatory and effector molecules in the initiation, progression, and resolution of periodontitis although the sum total of information on individual mediators is limited [
The periodontium is an intricate tissue in which hard and soft tissues are uniquely juxtaposed and which comprises numerous different cell types. The periodontium is exposed to a complex, variable, and chronic bacterial challenge, in which a chronic inflammatory process is often established. Thus, it is unsurprising that the profile of molecular mediators is similarly complex, dynamic, and variable. Also, in recent years it has become apparent that although we have a much improved understanding of fundamental cellular and molecular processes relevant to periodontal pathogenesis, these are often “snapshots” of individual pathways and our understanding of how these pathways are integrated and how they map to the initiation and progression of periodontitis and, critically, the possible resolution of this disorder is limited [
Clinical studies of GCF and saliva have identified a number of promising individual mediators which are associated with periodontal disease and correlated with clinical measurements of periodontitis, and whose levels change in parallel with the clinical course of the disease and in response to treatment [
The possible role of salivary proteins as biomarkers of periodontitis. “Robust” biomarkers are defined as those salivary proteins which have been shown to discriminate between periodontitis and oral health in at least 3 cross-sectional studies (with comparatively little or no published evidence to the contrary) and for which there may be supporting evidence from longitudinal studies investigating the natural course of periodontitis and/or the effects of treatment on biomarker levels. “Potential” biomarkers are identified using identical criteria to “robust” biomarkers with the exception that there are 2 replicated cross-sectional studies showing disease discrimination in addition to possible supporting evidence from longitudinal studies but for which there may be limited contradictory studies. It is accepted that the entries in the “robust” and “potential” categories may be interchangeable depending on the existence of further studies which remain unpublished for commercial reasons. “Uncertain” biomarkers are proteins for which there are only single studies showing discrimination of periodontitis or for which there are several studies from which the evidence is contradictory. “Unlikely” biomarkers are those proteins for which there are 3 or more studies which fail to provide evidence for an association with periodontitis in the absence of any evidence to the contrary. For a more detailed description of the published research studies and the putative role of these proteins in periodontitis see the main body of the text.
Evidence from studies of GCF suggests that cytokines likely form a substantial and measurable element of oral fluids [
Studies of salivary TNF-
The majority of studies of salivary IL-6 demonstrate no association with periodontitis [
Although there is a paucity of data on salivary chemokines in periodontitis, 2 studies have indicated that salivary CCL3 (MIP-1
Interestingly, there is evidence from 2 independent studies that levels of salivary soluble CD14 (sCD14) are elevated in periodontitis [
There is an emerging interest in a number of cytokines such as osteocalcin, RANKL, and OPG which are regulators of bone cell activity and, as such, mediate bone loss characteristic of periodontitis [
A number of studies of salivary biomarkers for periodontal disease have used candidate molecules identified through their established use as systemic markers of inflammatory disease in general medicine. For example, calprotectin is a neutrophil protein which is considered to be a marker of inflammation [
Saliva contains abundant antibacterial proteins such as lysozyme, and cystatins, but there is only a single report of significantly reduced levels of salivary cystatins in periodontitis [
Growth factors have diverse functions which overlap and contribute to immune responses but although elevated levels of growth factors including TGF-
Data from analysis of MMP in saliva in periodontitis are consistent with the findings from studies of GCF [
Salivary gelatinases (MMP-2 and MMP-9) and elastase have been found to be significantly elevated in periodontitis patients as compared to controls [
Collagen matrix degradation by proteases such as MMPs leads to the release of fragments of collagen into the circulation; these peptides can be assayed as a measure of bone resorption in periodontitis [
Aggressive (acute) periodontitis is a form of periodontal disease associated with rapid loss of alveolar bone and gingival attachment and as such has a distinct presentation and clinical course to the more common chronic periodontitis described above [
Although many markers are discriminatory, that is, exhibit significant alterations in periodontitis and robust, that is, have been found to be discriminatory in more than 3 independent studies, their clinical utility remains uncertain. Many studies outlined above have quantitatively correlated mediator levels with clinical measurements of periodontitis (e.g., bleeding on probing, clinical attachment loss, and pocket depth) although this data has not always been presented. However, it is unclear whether or not these measurements reflect current disease activity or are the result of historical disease progression or whether or not the levels of many candidate biomarkers significantly associated with periodontitis in cross-sectional studies accurately reflect the likely path of disease progression and response to treatment. Although a number of longitudinal studies have recorded changes in GCF levels of MMPs in parallel with the clinical course of periodontitis, the data for salivary analysis is rather limited and is the result of studies using clinical protocols which are very variable making meaningful comparisons between studies difficult to achieve. Therefore, the need for more substantial clinical studies very much still exists [
In broad terms, published longitudinal studies of salivary mediators in periodontitis may be categorised as studies of levels during the natural progression of the periodontitis in patients without therapeutic or healthcare interventions and studies of the effects of particular active healthcare regimens. In a prospective cohort study, 219 unselected subjects were monitored over a 4-year period, during which time they did not undergo any dental treatment and changes in periodontal pocket probing depth (as a measure of periodontitis) as compared to levels of salivary biomarkers and lifestyle factors; a multiple logistic model revealed that disease progression related to smoking habit and not the levels of any of some 9 protein biomarkers (IL-1
Numerous more studies have investigated the effects of various treatment regimens, including nonsurgical and surgical approaches, on salivary biomarkers in periodontitis. However, the treatment regimens applied have differed considerably between studies and, unfortunately, there are insufficient data to compare the effects of particular periodontal therapies on individual biomarkers. Nonetheless, a common finding is a decline in biomarker level in parallel with improvement in clinical measures of periodontitis. Thus, a significant reduction for active collagenase after surgical and nonsurgical treatment in both localised juvenile periodontitis and adult periodontitis patients has been recorded [
A key recent study reported the longitudinal clinical investigation of periodontitis in parallel to measurement of a panel of potential biomarkers, a cohort of 100 volunteers [
Given the complexity of the disease pathogenesis and progression and the recognised interindividual differences in disease experiences it is likely that multiple biomarkers will be required in order to provide full diagnostic efficacy [
It is possible that different tests will be applicable to screening for disease susceptibility as compared to those required for monitoring of therapy and treatment selection [
The majority of studies characterising salivary protein biomarkers in periodontitis published thus far have been based on markers derived from pre-existing information on the molecular pathogenesis of periodontitis; there needs to be a more open, unbiased approach to biomarker identification in periodontitis. Thus, high throughput proteomics holds promise for disease-associated biomarker identification [
A number of preliminary studies investigating the proteomic profile of saliva in periodontitis using a combination of 2D electrophoresis and mass spectrometry have been published [
Developments in bioinformatics will play a key role in the interrogation of data from proteomic studies, integration with other global analyses (e.g., transcriptomics and metabolomics), exploitation of knowledge of complex biochemical and cellular interactions, and ultimately, therefore, in the identification of salivary biomarkers not obvious from studies of candidate proteins [
Biomarkers must be straightforward to analyse; the readout should be readily interpretable, and give enhanced information such as being able to predict onset, measure activity, and monitor disease progression (e.g., from gingivitis to periodontitis or in terms of severity of periodontitis or relapse after treatment) and there is real promise that a number of the markers described in the previous section may fulfill these criteria. However, there are a number of scientific, clinical, and technological challenges to achieving the successful clinical application of salivary diagnostics in the management of periodontitis [
Despite our growing knowledge of the molecular mediators of periodontitis and the advent of salivomics, there remains a lack of substantial, well-controlled, clinical studies aimed at characterising the efficacy of individual biomarkers or combinations in monitoring periodontitis. In particular, the majority of such studies published thus far have been conducted on a cross-sectional basis rather than longitudinal design and another weakness of many studies is that they are statistically underpowered and focused on individual mediators [
A major issue in clinical studies of periodontal disease is cigarette smoking which modifies numerous salivary mediators [
In addition to a broadening of the biochemical analysis for biomarker into the realm of salivomics and more appropriate clinical validation studies for salivary biomarkers outlined above, there are a number of practical issues that need to be addressed before this approach is translated into a bona fide diagnostic test.
In particular, there needs to be a reliable method of salivary collection; to date, a great variety of approaches have been used which will impact on the effectiveness of the test [
The method of analysis needs to be fit for the purpose. Although most clinical studies of salivary protein biomarkers have employed conventional ELISAs, there has been some interest in developing novel point-of-care (POC) devices which may be used in the dental clinic and eventually domestically. The real promise of salivary analysis use is the ability of the patient or clinician to directly and continuously assess disease status, progression, and therapeutic efficacy. Sensitive analysis may even allow presymptomatic diagnosis. Thus, microfluidic and microelectromechanical systems (MEMS) and nanoelectrofluidic systems (NEMS) for salivary diagnostics have been under development for at least the last decade [
There needs to be an understanding of the issues relating to acceptance of these novel diagnostic approaches by dental clinicians and their patients alike. The derivation of the so-called “value propositions” is likely to be a useful approach whereby key questions such as feasibility, diagnostic accuracy, impact on patient management and outcomes, and also impact on society (including return on investment issues) are addressed [
We now have substantial information suggesting that a limited number of protein biomarkers may be efficacious in diagnosis and management of periodontitis. These markers (e.g., MMP-8 and IL-1
There has been some considerable interest in promoting the dental clinic as a venue for general medical diagnosis through physiological measurements (such as weight, height, and blood pressure), finger stick blood tests, and salivary diagnostics [
The author declares that he has, in the past, received funding for his research from Philips Oral Healthcare.
The author gratefully acknowledges his colleagues Dr. Katrin Jaedicke, Professor Philip Preshaw, and Miss Rachel Williams for their critical review of the paper.