Definitions used for a healthy periodontium—A systematic review

Abstract Objective To investigate the explicitness and variability of the definition of periodontal health in the current scientific literature. Material and methods The authors conducted a systematic literature review using PubMed and CENTRAL (2013‐01/2019‐05) according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines and the guidelines of the Meta‐analysis Of Observational Studies in Epidemiology (MOOSE) statement. Results A total of 51 papers met the predefined inclusion criteria. Of these, 13 papers did not report any explicit definitions of periodontal health. Out of the 38 remaining articles, half of them used a reference to support their definition and half of them not. The studies published in periodontics‐related journals or those that scored a low risk of bias for the methodical quality presented more explicit and valid definitions. Probing pocket depth was the most frequently used individual parameter for defining periodontal health. However, there were substantial variations in the methods of measurement and cut‐off values. Conclusions Given the diversity of periodontal health definitions, a cross‐study comparison is difficult. The results of this review may be useful in making others aware of the significance of standardizing the definition of a healthy periodontium.


| INTRODUC TI ON
The main objective of periodontal care is to reach and maintain a healthy periodontium. The definition of periodontal health plays a crucial role in population surveillance and the determination of critical therapeutic targets for clinicians. 1 Most studies traditionally regarded that a healthy periodontium is the opposite of case definitions of periodontal disease, as does the World Health Organization (WHO) defining health as an absence of illness. 2 Specifically, periodontal health refers to a state free from inflammation and characterized by shallow pockets and the absence of gingival bleeding. 3 However, there are a variety of case definitions, [4][5][6] and these definitions refer to an array of clinical signs and symptoms, such as probing pocket depth (PPD), clinical attachment loss (CAL) and bleeding on probing (BOP). 7 Consequently, we assume that there is heterogeneity in the definitions of periodontal health. The definition of periodontal health should be consistent, facilitating comparison of clinical studies. 8 Periodontal health was recently defined as the absence of clinically detected inflammation by the 2018 World Workshop of the European Federation of Periodontology (EFP) and the American Academy of Periodontology (AAP). 9 This EFP/ AAP definition is mainly based on PPD and BOP scores. To date, no overview of periodontal health definitions has been conducted. Therefore, this systematic review (SR) investigates the current scientific literature related to the definition of periodontal health.

| Protocol development
The protocol for this SR was developed "a priori," following an initial discussion among members of the research team according to the Cochrane Handbook for Systematic Reviews of Interventions and the guidelines of PRISMA and MOOSE. work that could meet the eligibility criteria of the study, so-called "snowball procedure." For details regarding the search terms used, see Appendix S1.

| Eligibility criteria
Publications were included only when they (a) were original studies, (b) were conducted in a human population, (c) were published in English, (d) contained a defined group of periodontal health or a non-defined control group as an opposite to the defined periodontal disease, and (e) their definitions described measurements and identified thresholds.

| Screening and selection
Two reviewers (AL and RZT) screened the titles and abstracts of the studies obtained during the search for eligible papers independently.
After the screening, the reviewers read the full texts of eligible papers in detail. Any disagreement concerning eligibility was resolved by consensus, and if conflict persisted, the decision was settled through arbitration led by a third reviewer (DES). The papers that met all the selection criteria were processed for data extraction.

| Assessment of heterogeneity
Heterogeneity across studies was detailed according to the following factors: study design, published journal type, subject characteristics, potential confounding factors, measurement tools and procedures, the number of explicit definitions, clinical parameters and cut-off values.

| Methodological assessment of risk of bias
The two reviewers independently scored the methodological qualities of each study as well (AL and RZT). The appropriate critical appraisal checklists from the Joanna Briggs Institute were used depending on the study design of the paper. 10 Studies that met 80% of the criteria were considered to have a low risk of bias. And 60% to 79% was a moderate one; 40% to 59% criteria were substantial one; and less than 40%, high one. 11

| Data extraction and analysis
The characteristics of the published journal type, study design, country, sample frame, sample size, group, age, gender, smoking status, medical condition, examination area, measurement tool, probing location and definition of periodontal health were extracted. Papers that included detailed measuring parameters and clear cut-off values were regarded as having an "explicit definition". 12 Moreover, the "explicit definition" papers that used references to support their definitions were viewed as having a "valid definition". 13 The extracted criteria for periodontal health were recorded with Microsoft Excel 2017 (Microsoft). All quantitative analyses were conducted with SPSS Statistics 25 (SPSS Inc).

| Search results
The search through online databases resulted in 1236 unique studies ( Figure 1). The initial screening of the titles and abstract resulted in 49 studies that went on to full-text review. Then, a detailed reading of the full texts was performed. Two independent reviewers excluded 20 studies (Appendix S2), leaving 29 eligible papers. Thirteen studies did not report an explicit periodontal health definition and rather referenced periodontal health as the opposite of disease. This study outlines the characteristics of the included papers.
These characteristics are summarized in Table 1.

| Methodological quality assessment
The methodological quality of the included studies was used to estimate the potential risk of bias and is presented in detail in Appendices S5.1-5. The estimated potential risk of bias was low for 15 studies, moderate for 25 studies, substantial for eight studies, and high for three studies.
The studies were designed as cross-sectional studies (37/51), longitudinal studies (4/51), and randomized or non-randomized allocated control studies (10/51). A total number of 372 983 individuals were enrolled in the studies, ranging from 18 to 354 850 individuals for each F I G U R E 1 Flow of information through the different phases of the systematic review. *see Appendix S2, **see Appendices S3 and S4, *** see

| Measurement methods
In 39 out of the 51 papers, a full-mouth assessment was conducted ( Moreover, five studies specifically measured the indicators at the location of the inter-proximal sites.

| Presence of an explicit or valid definition according to journal type, study design and risk of bias
A precise definition of periodontal health is offered in 38 (75%) of the included studies. The remaining 13 papers provided the references and defined the opposite of disease as periodontal health (Table 2).
An explicit definition with a supporting reference was reported in 19 papers. In contrast, the other 19 studies only used a definition rather than indicating any reference (Table 2; for details, see Appendix S6).
The two most frequently used references were the Armitage classification (1999), 17  In the periodontal journals, the definitions used were more explicit (87%) than those used in the dental or medical journals (Appendix S8.1). Moreover, the papers collected from a department of periodontology tended to provide explicit definitions (91%) compared with other studies. The studies scoring a low risk of bias for the methodical quality had more valid definitions (Appendix S8.2).  The most frequently used PPD cut-off was ≤3 mm, which appeared in 20 studies. However, 11 studies reported a threshold of 3.5 mm or higher. A considerable amount of variety was observed concerning the CAL threshold, ranging from 0 to 4 mm. Nine studies reported the absence of CAL, and 15 studies did not report CAL ( Figure 2B). Figure 2C demonstrates that among the reported BOP thresholds, the most commonly used was 10% sites, but the majority of the included papers (n = 23) did not report BOP. A periodontal pocket is the most common sign of periodontitis and easy to detect and assess in the clinical practice using various periodontal probes. The regularization of using periodontal probes will raise the accuracy of the process of diagnosing the condition and evaluating the treatment outcome. 19,20 The present SR has identified a great amount of variety in the methods and materials used, such as the periodontal probing methods, particularly the type of probe and probing site. The procedure of measuring PPD and CAL was described as being assessed by either four or six sites per tooth.

| D ISCUSS I ON
The number of sites used and especially the proportion of interdental sites assessed may influence the outcome. In any case, uniformity in material and methods can reduce the measurement bias.

The EFP/AAP workshop recommended the use of an International
Organization for Standardization (ISO) periodontal probe. 9 A cut-off or a reference point is needed to distinguish health from recurring signs and symptoms of periodontal disease. 8 A wide range of parameters and cut-offs were identified in the present systematic review. Probing pocket depth was the most frequently used periodontal parameter. Given the fact that it is rather easy to detect and measure, PPD has been recognized for many years as the essential parameter for the diagnosis of periodontal health and disease. 21 Half of the studies (51%) reported a threshold of PPD ≤3 mm. This cut-off value is also used to identify periodontal case types of health. 22 In contrast, there were still 11 (29%) definitions that used the threshold  The "only definition" and "reference and definition" groups were regarded as explicit definitions of periodontal health. b The "reference and definition" group was regarded as a valid definition of periodontal health. in large epidemiology studies stretch the PPD cut-off point. For instance, Hugoson used the following cut-off of periodontal health and disease: ≤10% sites with PPD ≥4 mm. 23,24 Nevertheless, even the largest cut-off value of PPD did not exceed 5 mm in the current review. A systematic review reported that probing depth up to 6 mm or even more should be taken into account as a high-risk factor to predict further disease progression in periodontal patients. 25 Other frequently used parameters are CAL and BOP. Clinical attachment loss, the second most frequently used parameter, varies across studies. This was used in three (8%) studies as the single parameter and in 21 (55%) as an adjunct to PPD. The most commonly used threshold using CAL is the absence of attachment loss.
As ageing comes with natural bone loss, some CAL is physiological.
Therefore, the absence of CAL is likely due to the outdated concept.
Periodontal health is identified as the absence of any deficit of supporting tissues. 8  Periodontal health can also present in an anatomically reduced periodontium. 1 In other words, periodontal health does not merely mean that there is an absence of supporting tissue deficit. It also refers to an individual's level of comfort, the stability of a functioning periodontium, and one's psychological and social well-being. This concept of holistic periodontal health has not been taken into account in this paper. Notably, the feasibility of directly regarding the definition of periodontal health in a reduced periodontium (PPD ≤4 mm and BOP ≤10%) as the treatment goal among patients remains uncertain.
However, a certain PPD value after treatment needs to be interpreted in the light of variance in susceptibility and personalized medicine.
Lang and Tonetti built a functional diagram to assess periodontal risk in supportive periodontal therapy, which can help clinicians distinguish whether a treatment goal is reached or not. 32 Moreover, the number of residual pockets with a probing depth of ≥5 mm to a certain extent reflects the degree of success of periodontal treatment, which is different from the PPD threshold in the new definition. In the randomized clinical trial, 33 the subjects presenting ≤4 sites with PD ≥5 mm at one year represented a successful treatment outcome.
Therefore, the endpoint of therapy should seek the most optimal balance between over-and underestimation of health status among treated periodontal patients. It is important to acknowledge the distinction between the diagnoses of periodontal health of initial patients versus treated patients. For the latter, a more flexible, comprehensive and detailed assessment would be recommended.

| CON CLUS ION
This SR revealed a variety of definitions of periodontal health in existing scientific literature. This heterogeneity was measured according to study characteristics, measurement methods, explicit definitions, references and cut-off values used. The definition of periodontal health proposed by the EFP/AAP Workshop offers an opportunity for the field to standardize and achieve uniformity in terms of methodologies in order to draw comparisons between different studies. This study also revealed that the number of people thought to have periodontal disease is likely overestimated due to the strict cut-off value.

| Scientific rationale for the study
There is no standard reference for periodontal health, and the diagnostic properties of the various definitions have not been studied.

| Principal finding
Marked heterogeneity in the definitions of different measuring methods and clinical parameters in periodontal health may be affecting interpretations of research.

| Practical implications
The new definition of periodontal health proposed by the EFP/ AAP workshop in 2018 offers an opportunity to standardize and unify the cut-off values of clinical parameters, which would allow for a better comparison of clinical studies and support research and decision-making.

ACK N OWLED G EM ENTS
The authors acknowledge the help of Dr Diane Black, lecturer of language centre of University Groningen, with proofreading.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
An Li, first author, contributed to the acquisition, analysis and interpretation of data, and drafted the manuscript. Renske Z. Thomas, overall daily supervisor, contributed to the design of study, the acquisition, analysis and interpretation of data, and drafted the manuscript.
Luc van der Sluis contributed to the design of study and critically revised the manuscript. Geerten-Has Tjakkes contributed to the design and critically revised the manuscript. Dagmar Else Slot contributed to the conception and design of the study, supported the analysis and interpretation of the data, and critically revised the manuscript. All authors gave final approval and agreed to be accountable for all aspects of this work, ensuring its integrity and accuracy.