1.Background

Dental plaque is attached to a solid surface, such as the enamel of teeth and the surface of root or dental implants, and plays a key role in the etiopathogenesis of oral diseases, such as dental caries and periodontal diseases^1,2). Therefore, the quantification of dental plaque is a critical issue for oral hygiene assessments and professional mechanical plaque control.

In 1950, Hine argued that the accumulation of oral debris leads to gingival inflammation³⁾. In addition, in 1959, Schei et al.⁴⁾ demonstrated the relationship between oral hygiene and alveolar bone loss using epidemiological data. The relationship between dental plaque and gingivitis was first established in 1965 by Löe et al.⁵⁾ who argued that the removal of dental plaque resolves gingival inflammation.

Clinically, dental plaque is critical because bacteria are the main causative and pathogenic factor⁶⁾. Hence, various methods have been developed to visually evaluate dental plaque in the oral cavity. Among these, observation of the dental plaque using a disclosing agent is the most traditional method, and the erythrosine-disclosing agent recommended by Arnim⁷⁾ in 1963 is the most widely used. By using a disclosing agent to stain dental plaque, which is difficult to evaluate visually, the area covered and the amount of dental plaque can be confirmed. Moreover, dentists can evaluate the oral condition of the patient based on the disclosed tooth surface.

Various dental plaque indices have been developed to quantify the amount of dental plaque accumulated on the tooth surface. The Patient Hygiene Performance Index (PHPI) was developed by Podshadley and Haley⁸⁾ in 1968. In this, each tooth surface is divided into five sections. If no disclosed tooth surface is present, a score of 0 is assigned to that section, and if a disclosed tooth surface is present, a score of 1 is assigned. The Quigley and Hein Plaque Index (QHPI) was developed by Quigley and Hein⁹⁾ in 1962 and partially modified by Turesky et al.¹⁰⁾ in 1970. In the QHPI, scores are assigned based on the width and accumulated quantity of dental plaque, and the amount of dental plaque is indicated by a score ranging from 0 to 5 points. Rustogi’s Modified Navy Plaque Index (RMNPI) was developed by Rustogi et al.¹¹⁾ in 1992; wherein, the tooth surface is divided into nine sections—six for the root and three for the crown—based on the center of the tooth. As with the PHPI, a score of 1 is assigned if a disclosed tooth surface is present and a score of 0 is assigned if no disclosed tooth surface is present.

The three previously described dental plaque indices are well established, and their reliability, validation, and reproducibility have been proven through various studies^12-16). However, the PHPI and RMNPI have a limitation in assigning a score of 1 when a small amount of dental plaque is present on each subdivided tooth surface, as it was assumed that the plaque covers the entire area. In addition, with the QHPI, assessing the quantity of dental plaque is difficult when the plaque is deposited in a noncontinuous manner from the gingival margin and instead accumulates on the proximal surface and crown. Since the deposition pattern of dental plaque varies greatly depending on the eruption status and oral environment, previously used dental plaque indices sometimes fail to fully represent the deposition pattern of the plaque.

2.Objectives

This study aimed to compare the existing dental plaque indices and values obtained through digital analysis and to develop a Novel Plaque Index (NPI) that is closely related to digital analysis values.

1.Subjects and samples

Ethical approval was obtained from Institutional Review Board of Pusan National University Dental Hospital (IRB No. PNUDH-2016-004). Ninety-six adult participants, aged 30∼65 years, were recruited. Before the clinical trial, the participants were given information about the purpose and methods of the study and asked to sign informed consent forms. The disclosing procedure was performed to evaluate the oral condition of the participants using a disclosing agent (Garnet Disclosing Solution; Dharma Research Inc., Miami, FL, USA), and the disclosed tooth surfaces were photographed using a Whicam Story3-wired type intraoral camera (GoodDrs, Incheon, Korea). Of the 96 participants, 40 were selected whose oral photography revealed dental plaque deposition, which was difficult to quantify using the existing dental plaque indices, were selected. The tooth surfaces analyzed in the study were the buccal surface of the maxillary left first molar and the labial surface of the mandibular left central incisor, with a total of 80 intraoral camera images (two images per participants). The 40 selected participants included 8 male and 32 female with a mean age of 41.5 years for the male and 40.3 years for female (Table 1) .

Table 1

Distribution of the Study Subjects by Sex and Age Group

Age	Total	Male			Female		p-value
		n	%		n	%
<40	21	5	23.8		16	76.2	0.698
≥40	19	3	15.8		16	84.2

p-value by chi-squared test.

2.Quantification of dental plaque by image analysis

The digital image analysis program ImageJ (National Institutes of Health, Bethesda, MD, USA) was used by an examiner to directly designate the entire tooth surface as well as areas with deposited dental plaque and to express the specified area as a number. It is a widely used representative image analysis program that is currently used in various types of research¹⁷⁾. In this study, ImageJ was used to analyze the images obtained by the intraoral camera and the amount of dental plaque deposited on the tooth surface was quantifies using the the Plaque Percent Index (PPI)¹⁸⁾. The PPI score calculated through image analysis is considered the gold standard. Assessment using the existing dental plaque indices was conducted through visual observation by examiners without the use of ImageJ.

3.Development of the Novel Plaque Index

The tooth surface was divided longitudinally into the mesial, middle, and distal thirds. Each longitudinal third was subdivided into four segments sections (half of the cervical one-third, cervical one-third, middle one-third, and incisal or occlusal one-third) based on the reference points. The reference points were as follows; the cross points between the tooth and interdental papilla and the highest contact points of the tooth (Fig. 1).

Fig. 1. Twelve subdivisions of a tooth surface used in the Novel Plaque Index.

A score of 0 to 5 was assigned to each longitudinal third, and the average score was determined based on the NPI score of the tooth surface. The scoring system is as follows:

0: no plaque

1: noncontinuous plaque at the cervical margin of the tooth

2: the presence of continuous plaque at the second section

3: the extension of continuous plaque to the third section from the bottom

4: the extension of continuous plaque to the fourth section from the bottom

5: the extension of continuous plaque to the fifth section from the bottom

Noncontinuous dental plaque from the gingival margin was excluded from the score calculation.

4.Quantification of dental plaque by visual observation

The examiners calculated the PHPI, RMNPI, QHPI, and NPI scores using images captured with an intraoral camera and a tablet PC (LG, Seoul, Korea). Dental plaque indices were calculated by two trained examiners, and each examination was repeated twice for reliability analysis. The second assessment was performed after a washout period of 2 weeks to prevent the results of the primary assessment from affecting those of the secondary assessment.

5.Statistical analysis

The relationships between the PPI and dental plaque indices were evaluated using Pearson’s correlation coefficient, and the reliability of intraexaminer and interexaminer agreement was tested using the intraclass correlation coefficient (ICC) (PASW Statistics ver. 23.0; SPSS, Chicago, IL, USA) (α=0.05) and weighted kappa (Microsoft Office Professional plus Excel 2016). When the reliability coefficient was below 0.40, the level of clinical significance was considered poor; when it was between 0.40 and 0.59, the level of clinical significance was considered fair; when it was between 0.60 and 0.74, the level of clinical significance was considered good; and when it was between 0.75 and 1.00, the level of clinical significance was considered excellent¹⁹⁾.

1.Validity analysis of the Novel Plaque Index

Correlation analysis was conducted to determine the relationships between the PPI score calculated via image analysis and the PHPI, QHPI, RMNPI, and NPI scores. The Pearson correlation analysis performed by examiners 1 and 2 revealed that the NPI was more closely related to the PPI compared to the existing dental plaque indices. For the anterior, posterior, and total teeth surfaces, the NPI correlation coefficients of examiner 1 were 0.84, 0.87, and 0.87, respectively, and those for examiner 2 were 0.87, 0.87, and 0.88, respectively (Table 2).

Table 2

Pearson’s Correlation Coefficient of the Plaque Percent Index with the Existing Dental Plaque Indices and the NPI

		PHPI	QHPI	RMNPI	NPI
Examiner 1	Anterior	0.82**	0.69**	0.71**	0.84**
	Posterior	0.62**	0.76**	0.85**	0.87**
	Total	0.75**	0.71**	0.78**	0.87**
Examiner 2	Anterior	0.70**	0.41**	0.70**	0.87**
	Posterior	0.49**	0.58**	0.61**	0.87**
	Total	0.64**	0.47**	0.69**	0.88**

PHPI: Patient Hygiene Performance Index, QHPI: Quigley and Hein Plaque Index, RMNPI: Rustogi’s Modified Navy Plaque Index, NPI: Novel Plaque Index.

**p<0.01.

2.Intraexaminer agreement of examiner 1

The ICC reliability analysis results for examiner 1 showed that the reliability of the NPI was higher than that of the existing dental plaque indices, and the reliability values were 0.97, 0.92, and 0.95 for the anterior, posterior, and total teeth surfaces, respectively. Moreover, the NPI reliability values in the weighted kappa results were 0.91, 0.83, and 0.87 for the anterior, posterior, and total teeth surfaces, respectively (Table 3).

Table 3

Results of Intraclass Correlation Coefficient and Weighted Kappa Reliability of Examiner 1

		PHPI	QHPI	RMNPI	NPI
Intraclass correlation coefficient	Anterior	0.93	0.88	0.91	0.97
	Posterior	0.84	0.88	0.91	0.92
	Total	0.93	0.88	0.89	0.95
Weighted kappa	Anterior	0.80	0.67	0.77	0.91
	Posterior	0.73	0.77	0.87	0.83
	Total	0.80	0.70	0.78	0.87

PHPI: Patient Hygiene Performance Index, QHPI: Quigley and Hein Plaque Index, RMNPI: Rustogi’s Modified Navy Plaque Index, NPI: Novel Plaque Index.

3.Intraexaminer agreement of examiner 2

The ICC reliability analysis results for of examiner 2 revealed that the NPI reliability value was 0.93 for the anterior, posterior, and total teeth surfaces. The NPI reliability values in the weighted kappa results were 0.87, 0.86, and 0.85 for the anterior, posterior, and total teeth suefaces, respectively (Table 4).

Table 4

Results of Intraclass Correlation Coefficient and Weighted Kappa Reliability of Examiner 2

		PHPI	QHPI	RMNPI	NPI
Intraclass correlation coefficient	Anterior	0.95	0.91	0.96	0.93
	Posterior	0.72	0.55	0.83	0.93
	Total	0.94	0.77	0.94	0.93
Weighted kappa	Anterior	0.93	0.97	0.92	0.87
	Posterior	0.54	0.79	0.85	0.86
	Total	0.87	0.63	0.85	0.85

PHPI: Patient Hygiene Performance Index, QHPI: Quigley and Hein Plaque Index, RMNPI: Rustogi’s Modified Navy Plaque Index, NPI: Novel Plaque Index.

4.Interexaminer agreement

Among the indices, the NPI demonstrated the highest ICC reliability. The reliability values were 0.95, 0.90, and 0.93 for the anterior, posterior, and total teeth surfaces, respectively. The weighted kappa values calculated by the two examiners were 0.87, 0.76, and 0.81 for the anterior, posterior, and total teeth surfaces, respectively (Table 5).

Table 5

Results of Interclass Correlation Coefficient and Weighted Kappa Reliability of Examiners 1 and 2

		PHPI	QHPI	RMNPI	NPI
Intraclass correlation coefficient	Anterior	0.89	0.45	0.90	0.95
	Posterior	0.73	0.62	0.78	0.90
	Total	0.89	0.49	0.86	0.93
Weighted kappa	Anterior	0.65	0.30	0.79	0.87
	Posterior	0.58	0.20	0.63	0.76
	Total	0.67	0.15	0.67	0.81

PHPI: Patient Hygiene Performance Index, QHPI: Quigley and Hein Plaque Index, RMNPI: Rustogi’s Modified Navy Plaque Index, NPI: Novel Plaque Index.

1.Interpretation and comparison with previous studies

In this study, we evaluated the relationship between the amount of dental plaque determined through image analysis (using the PPI), existing dental plaque indices, and the NPI using intraoral camera images. A reliability analysis of the calculated values was performed by two examiners using the ICC and weighted kappa values. The study results revealed that the NPI was more closely related to accumulated dental plaque than the existing dental plaque indices. Additionally, the NPI reliability agreement was high, according to the ICC and weighted kappa values obtained in the reliability analysis.

In a previous study, Carter et al.²⁰⁾ photographed disclosed tooth surfaces using a camera and measured the amount of dental plaque using a fully automated method. In addition, Carvalho et al.²¹⁾ evaluated the validity of the Visible Occlusal Plaque Index for assessing occlusal surfaces. Various dental plaque indices and methods have been developed, and the NPI demonstrated high precision in quantifying plaque, with excellent reliability. In this study, measurements were repeated by each examiner to differentiate the results from those of previous studies, and the ICC and weighted kappa values were calculated to estimate intraexaminer and interexaminer reliability. Because the NPI has established reference points in the periodontal tissue and adjacent teeth, the shape of the dividing line can be vary depending on the patient’s oral condition. Three existing dental plaque indices were used to evaluate the disclosed teeth, with an emphasis on tooth segmentation rather than the patient’s oral condition. For example, when comparing patients with and without gingival recession or interdental papilla loss, the removal of dental plaque on the proximal surface is easier for those with interdental papilla loss. Nevertheless, if dental plaque on the proximal surface is not removed, it should be evaluated more strictly (Fig. 2). Hence, the NPI has established a new standard for dental plaque indices. In addition, the division of the tooth surface into three sections and exclusion of noncontinuous dental plaque from the gingival margin were aimed to develop a dental plaque index capable of calculating the proportional amount of deposited dental plaque.

Fig. 2. Example of subdivisions of tooth surface in different periodontal status.

Although the amount of deposited dental plaque covered approximately 33% of the tooth surface, the scores obtained using the existing dental plaque indices were the highest (Fig. 3). Although the NPI also yielded high scores of 4 and 3.67 points, the amount of dental plaque deposited on the tooth surface was more clearly expressed than that in the existing dental plaque indices.

Fig. 3. The score of Plaque Percent Index (PPI), Patient Hygiene Performance Index (PHPI), Quigley and Hein Plaque Index (QHPI), and Rustogi’s modification Modified Navy Plaque Index (RMNPI) and Novel Plaque Index (NPI) for the two tooth surface images.

In this study, we used weighted kappa values instead of Cohen’s kappa values for the reliability analysis²²⁾. Although Cohen’s kappa²³⁾ is a commonly used method for reliability analysis used in two categories, it can be used for multiple categories, though this reduce its reliability value²⁴⁾. In this case, partial agreement can be recognized when a judgement is close to a certain category. Hence, the weighted kappa carries more weight than the Cohen’s kappa in the reliability analysis²⁵⁾. Because the PHPI, QHPI, and NPI have six categories, we assigned a weight of −0.2 per point. For the RMNPI, Which has 10 selectable categories, we assigned a weight of –0.12 per point.

2.Limitations

Capturing a clear image of the tooth surface using an intraoral camera was crucial for conducting this study appropriately. If distortion had occurred in the area to be analyzed, it would have been superimposed on the proximal or occlusal surface, making quantification of the deposited dental plaque difficult. Additionally, drawing a dividing line on the tooth surface is difficult if adjacent teeth are missing. In such a case, the opposite tooth and the contact point should be assessed. Moreover, it was difficult to assign a score simultaneously because of the large number of subdivided tooth surfaces, necessary to accurately assess the amount of dental plaque. In other words, the NPI may be less advantageous than traditional dental plaque indices because it divides the tooth into multiple surfaces, making simple visual inspection more complex. However, the NPI effectively reflects the amount of dental plaque using an image analysis program, which has improved the accuracy of the assessment. Finally, an intraoral camera is necessary to calculate the index; however, as the advancements in imaging equipment continue, it is reasonable to incorporate it into research. Despite some limitations, the NPI, which is closely related to the amount of dental plaque deposited on dental biofilms, showed excellent validity and reliability. It accurately reflects the deposition pattern of dental plaque and will be helpful for examining the oral cavity of patients in research and clinical applications. In future studies, it will be essential to calculate the index score, including not only the anterior and posterior teeth but also additional tooth surfaces, including lingual surfaces.

3.Conclusion

We confirmed that the NPI proposed in this study more clearly represents dental plaque accumulation than the existing dental plaque indices and offers excellent reproducibility. The NPI will be more useful than the existing dental plaque indices for calculating the amount of dental plaque on disclosed tooth surfaces in research and clinical settings.

None.

No potential conflict of interest relevant to this article was reported.

This study was approved by the institutional review board of Pusan National University Dental Hospital (IRB No. PNUDH-2016-004).

This work was supported by Youngsan University Research Fund of 2024.

Raw data is provided at the request of the corresponding author for reasonable reason.