Participants
In this cross-sectional study, oral health in a 65-year-old population in Norway was investigated. A random sample was drawn from the Norwegian Population Register (retrieved from the Norwegian Tax Administration) and invitation letters were sent out. Inclusion criteria were “born in 1954” and “resident in Oslo”. All individuals who received the letter and were reachable by phone were contacted and given the opportunity to participate in the study. The recruitment procedure has been described in detail in a previous publication [21]. The study was approved by the Norwegian Regional Committee for Medical and Health Research Ethics (REK 2018/1383) and performed in compliance with the tenets of the Declaration of Helsinki. A written informed consent was signed by each participant prior to the clinical examination.
Questionnaire
All participants answered a semi-structured questionnaire prior to the clinical examination using the Nettskjema software (University of Oslo, Norway). The questionnaire contained items regarding general diseases, medication use, and smoking habits. Self-reported diabetes type 1 and 2 were assessed by yes/no questions. Smoking habits were assessed by the three response alternatives: “never smoker”, “former smoker”, and “current smoker”. “Current smoker” was defined as an individual who smoked at least one cigarette daily. Current smokers also reported the number of cigarettes daily consumed. The participants’ country of birth was dichotomized into ‘western’ (Nordic countries, Western Europe, North America, and Australia) and ‘non-western’ (the rest of the world). The level of education was dichotomized into ‘higher education’ (university/college education) and ‘basic education’ (high school, elementary school, or lower).
Clinical periodontal examination
Two trained, calibrated dentists (ATTS and MTD) performed all clinical examinations at the Research Clinic at the Faculty of Dentistry, University of Oslo from February to December 2019. Periodontal probing depths (PPD), bleeding on probing (BoP) and suppuration were measured on six sites per tooth using LM 52B XSI Perio Probe (LM-Dental™, Planmeca Group, Helsinki, Finland). PPD was recorded to the nearest mm, rounded down. Tooth mobility was recorded as grade 1, 2, or 3 as described by Nyman et al. [22]. Furcation involvement was measured on molars using Nabers Q2N probe (Hu-Friedy, Chicago, USA) and recorded as grade 0, I, II, or III [23]. Third molars were not included in the clinical registrations.
The examiners were trained by a specialist in periodontology (OCK) prior to inter-rater calibration and clinical examination. Inter-rater reliability of the clinical examiners was estimated from pocket depth registrations from seven participants, a total of 336 values per examiner. The intra-class correlation coefficient (ICC) (95% CI) was 0.82 (0.78–0.86).
Radiographic examination
Orthopantograms (OPG) and horizontal bitewings (BW) were used to assess radiographic bone loss (RBL). The OPGs were obtained using a panoramic imaging unit (ProMax X-ray Dimax 3 and Planmeca ProOne, Planmeca Oy, Helsinki). Two BWs per participant were obtained using an intraoral imaging unit (MINIRAY, SOREDEX, PaloDEx Group Oy, Tuusula, Finland) with a rectangular collimator (length 30.5 cm). Percentage bone loss was calculated for each tooth by dividing the distance from cemento-enamel junction to alveolar crest by the distance from the cemento-enamel junction to apex measured on OPGs. In addition, detectable interproximal bone loss was recorded as the distance in millimeters from the cemento-enamel junction to the alveolar bone crest on the two most severely affected non-adjacent sites on BW.
Radiographic registrations were performed by one calibrated dentist (ATTS) in a room with adapted ambient light and by applying measuring instruments in the ImageJ software (ImageJ 1.52a, National Institutes of Health, USA). Calibrations were performed on OPGs from 50 participants. A specialist in periodontology (OCK) and a trained dentist (ATTS) separately evaluated 50 radiographs. The results were evaluated and for the inter-examiner calibration the ICC (95% CI) was 0.79 (0.66–0.86) for % bone loss and the weighted Cohen’s kappa (95% CI) was 0.72 (0.66–0.78) for stage. Intra-examiner calibration was performed (ATTS) on 25 radiographs. For the intra-examiner radiographic registration the ICC (95% CI) was 0.88 (0.86–0.90) for % bone loss and the weighted Cohen’s kappa (95% CI) was 0.90 (0.82–0.98) for stage.
Definitions
The results from the clinical and radiographic examination in the present study were classified based on the 2017 World Workshop on the Classification of Periodontal and Peri-implant Diseases and Conditions [19, 24, 25]. A periodontitis case was defined as an individual with detectible RBL (distance from the cemento-enamel junction (CEJ) to the alveolar bone crest (AC) measured on BW radiographs) exceeding 3 mm on ≥ 2 non-adjacent teeth. For further classification of disease activity, the following definitions based on Sanz et al. [26] were used:
-
Periodontitis case, stable case of periodontal health: PPD ≤ 4 mm, BoP < 10%, no PPD ≥ 4 mm with BoP
-
Periodontitis case, some gingival inflammation: PPD ≤ 4 mm, BoP ≥ 10%, no PPD = 4 mm with BoP
-
Periodontitis case, unstable case of recurrent periodontitis: PPD ≥ 5 mm, or PPD ≥ 4 mm with BoP
Percentage of radiographic bone loss measured on each tooth was used for evaluating the severity of periodontitis according to Papapanou et al. [19]: < 15%; stage I, 15–33%; stage II, extending to mid-third of root and beyond; stage III. In addition, PPD, furcation involvement, and tooth loss due to periodontitis were investigated as complexity factors and used for further staging of the periodontitis cases as described by Papapanou et al. [19]. Vertical bone loss was registered if the radiographic defect was ≥ 3 mm deep and ≤ 3 mm wide [6]. Teeth were registered as lost due to periodontitis based on clinical judgment if this was considered as the most likely cause for tooth loss, that is, cases where no other reasons for tooth loss such as caries, endodontic lesions, root canal treatments seemed reasonable, and the general bone level in remaining dentition suggested periodontal tooth loss. In cases of doubt, tooth loss was not assigned to periodontitis. For each stage group the extent was described as localized (< 30% of teeth involved) or generalized (≥ 30% of teeth involved).
For grading of periodontitis cases, percent bone loss at the most affected tooth divided by age was calculated and categorized as grade A (< 0.25), B (0.25–1.0), or C (> 1.0). In addition, smoking and diabetes were treated as grade modifiers. The grade modifiers could only shift the grade, based on radiographic bone loss, to a higher level. Smokers who smoked < 10 cigarettes daily were categorized as grade B and individuals who smoked ≥ 10 cigarettes daily were categorized as grade C. In lack of hemoglobin A1c measurements, diabetes could only shift the grade from A to B and not from B to C.
Staging and grading were only performed on periodontitis cases. For non-periodontitis cases, the following definitions based on Chapple et al. [24] were used:
In cases with no detectable bone loss, PPD > 3 mm and BoP < 10%, pocket depths were assigned to anatomical causes, and the participants were defined as cases with periodontal health. In cases with PPD > 3 mm, no detectable bone loss on radiographs, and BoP ≥ 10%, pocket depths were considered as pseudo pockets and the participants were assigned to the gingivitis group. Gingivitis was defined as localized when BoP was present in < 10% of sites and generalized when BoP was present in ≥ 10% of sites.
In addition to the 3 mm interproximal RBL cutoff used for identifying periodontitis cases, the difference in prevalence, stages, and grades when using RBL > 1 mm and RBL > 2 mm cutoffs were investigated.
Statistical analyses
Clinical and radiographic registrations were collected in The Oral Data Collector sheet specifically designed for data entry in this study, developed in Microsoft Excel 2016 (Microsoft Corporation, Redmond, Washington, US), and imported into STATA (Stata version 16.1; College Station, TX, USA) for statistical analysis. Data were stored in Service for Sensitive Data (TSD facilities, UiO). Participants with ≤ 1 remaining tooth were excluded from the analyses. The results from the descriptive analyses are presented as percentage distributions or mean and standard deviation (SD).