Introduction

The concept of deep margin elevation (DME), also known as cervical margin relocation, proximal margin elevation, and proximal box reduction (PBR), is well established [1, 2]. DME aims to relocate the interproximal subgingival margin of an indirect restoration to a more biologically favorable location (supragingival or equigingival) to avoid the need for surgical crown lengthening [3, 4]. The margins of indirect metal or ceramic restorations should be placed on healthy tooth structures [5]. As large carious lesions may extend subgingivally, periodontal health can be compromised following the introduction of restorative material or an indirect restoration into the periodontium. If the direct or indirect restorative material is placed into the epithelial attachment, violation of the biological width might affect the general health of the periodontal apparatus, leading to bone resorption [6].

Adequate space for restorative margin placement can be achieved through surgical (crown lengthening) or orthodontic treatment (forced eruption) [7]. Although relatively cost-effective, crown lengthening is a more invasive procedure that requires a prolonged healing time, especially in the esthetic zone. As it is a surgical procedure, it has several associated negative sequelae including post-operative discomfort and bleeding, sensitivity, formation of black triangles, and loss of interdental papillae. By contrast, forced eruption is a non-surgical treatment option that modifies the osseous and gingival topography. While a conservative treatment, it is both expensive and time-consuming [8].

Deep carious lesions extending beyond the cementoenamel junction are common, occurring in 36 to 67% of the young population and 88.4% of elderly patients. They most commonly occur in the canines followed by the premolars. Hence, there is a need for minimally-invasive, time-efficient, and patient-friendly techniques for DME [9].

Indirect ceramic restorations have predictable long-term outcomes and superior esthetics and mechanical properties, especially when bonded to enamel. However, moisture control is mandatory for a predictable restoration-tooth bond. Thus, elevating the deep margins with direct composite facilitates bonding to an accessible supragingival margin compared with bonding to a non-raised subgingival margin, which has a negative impact on bond strength, plaque accumulation, and the surrounding gingival tissue [10].

Although DME is considered less invasive than surgical crown lengthening, it is still technically demanding. Clinicians may find it difficult to create proper anatomical contours at subgingival levels, as finding a matrix reaching subgingival areas is challenging. It has been suggested that DME should be a two-step procedure, first to provide a cervical seal and anatomical emergence profile, and second to complete the restoration with optimal contact areas. Marginal adaptation has been studied in vitro by measuring the gaps created between the tooth tissue and restorations using scanning electron microscopy [11]. Micro leakage has also been assessed in vitro using dye-penetrating methods, which revealed no significant difference between DME and non-DME approaches [12].

Margins can be raised using either resin-modified glass ionomer (RM-GIC) and/or flowable composite. RM-GIC releases fluoride but has a high dissolution rate, while composite may better bind to teeth and ceramic with less microleakage and increased fracture resistance. Indeed, increased fracture resistance with flowable composite as a liner below class II amalgam restorations has been demonstrated in vitro [13]. Flowable composite with inferior mechanical properties beneath composite resin acts as a shock absorber and may improve sealing and marginal adaptation. Microleakage has been shown to be reduced by delayed light curing or soft-start polymerization [14]. Several in vitro studies have demonstrated that marginal adaptation of indirect inlay restorations was similar with and without DME [15].

However, there is currently no information on levels of dental practitioner knowledge about DME. To fill this gap, here we assessed the (i) the level of knowledge and opinions of periodontists, prosthodontists, restorative dentists, and general dentists practicing in Saudi Arabia dental hospitals and private practices about DME, and (ii) associations between levels of knowledge and gender, dental specialty, years of clinical experience, practice setting (academic vs clinical), and country of clinical training. The null hypothesis was that dentists and dental specialists are equally knowledgeable about DME and advocate the use of this concept in their daily practice.

Materials and Methods

Sample

This was a study of 432 dentists working in Jeddah, Saudi Arabia. Ethical approval was obtained from the Research Ethics Committee, Faculty of Dentistry, King Abdulaziz University (REC No. 10-12-19). The dental personnel included in the study were undergraduate students, general dentists, advanced general dentists, periodontists, restorative dentists, or prosthodontists practicing in public or private hospitals/clinics in Saudi Arabia. We excluded pedodontists, orthodontists, endodontists, maxillofacial surgeons, and non-practicing dentists and dental specialists. At the start of the survey, the main purpose, a summary of the study parts, and the time required to complete the survey were provided. Once the participant had read this information and accepted it, he/she signed a consent form before completing the survey.

Questionnaire validity and reliability testing

A questionnaire was developed in accordance with the research objectives and was divided into two main parts. The first contained demographic data including gender, educational degree, country of clinical training, work setting, years of clinical experience, and specialty. The second part contained questions about knowledge and its source.

The questionnaire was tested for face and content validity. Twenty specialists and general dentists with expertise in DME were interviewed and asked to complete the survey and evaluate the level of importance and relevance of each question on a three-point scale (important, neutral, not important). Their opinion on whether a question should be included or not was also evaluated using a yes or no question format. Moreover, the clarity of each question was discussed with each evaluator. Twenty specialists were asked to complete the survey after ten days to assess reliability. Kappa statistics were calculated, and the survey was adjusted according to the results.

Survey distribution

The survey included 22 questions with sub-questions. The estimated time to complete the survey was 8-10 minutes. Some subjects were approached randomly in-person to complete the survey using an electronic tablet, while others received the survey by email using the Google Forms platform. Google Forms was chosen as it is user-friendly, free of charge, easily accessible on different electronic devices, can be sent via email and different applications, and provides the option of output to Excel spreadsheets or SPSS software.

Statistical analysis

Data were input into IBM SPSS Statistics version 20 (IBM Corp., Armonk, NY) and analyzed for descriptive statistics (frequencies). Differences between the two groups were calculated using the chi-squared test or Fisher’s exact test. Binary regression was performed to assess predictors of DME knowledge. P-values <0.05 were considered significant.

Results and Discussion

Sample characteristics

Post-hoc calculation of the power of the chi-square test was performed using G*Power software. For α = .05, an effect size of 0.3, a sample size of 432, and a maximum df = 5, power was calculated as 0.999. Counts and percentages, and Chi-square tests were shown in Tables 1-4. Post-hoc calculation of the power for binary logistic regression was also performed using G*Power software, (Tables 2-4). For α = .05, a sample size of 432, and a two-tailed normal distribution, the power was calculated for each odds ratio mentioned in each characteristic. Binary logistic regression tests (Tables 2-4).

There were 432 responses, of which 51.6% were from males and 48.4% were from females. Forty-nine percent had postgraduate training and 60% were clinically trained in Arabian or Asian countries. Half of the participants (50.8%) worked both as academics and clinicians. Forty-six percent of the participants had work experience of 1-10 years, while only 18.5% had work experience of over 10 years. Forty-nine percent of participants were general dentists followed by 22.5% restorative/advanced general dentists (Table 1).

Table 1. Characteristics of the respondents.

AGD: Advanced General Dentists

Knowledge about DME

Eighty-five percent of the cohort knew about the concept of DME. In univariable analysis, gender, work setting, and years of experience were associated with knowledge about DME (chi-squared test, p<0.05). However, in multivariable analysis, only work setting (odds ratio (OR) 2.2, 95% CI 1.1-4.3; p<0.05) and 0 years of experience (OR 20.9, 95% CI 4.9-88.2; p<0.001) were associated with knowledge about DME (Table 2). 

Table 2. Effect of participant’s gender, education, training country, practice setting, years of experience, and specialty on knowledge about DME and its predictors.

Source of knowledge about DME

Fifty percent of the cohort used the three assessed sources (colleagues, social media, and the scientific literature) to obtain information about DME. In univariable analysis, all assessed parameters (educational attainment, training country, work setting, experience, and specialty) except gender were associated with sourcing information about DME (chi-squared test, p<0.01). Consistently, in multivariable analysis, educational degree (OR 0.4, 95% CI 0.3-0.6; p<0.001), training country (OR 0.5, 95% CI 0.3-0.8; p<0.01), work setting (OR 0.3, 95% CI 0.2-0.5; p<0.001), 0 years of experience (OR 3.6, 95% CI 1.7-7.5; p<0.01), 1-10 years of experience (OR 12.5, 95% CI 6.3-24.8; p<0.001), periodontists (OR 3.5, 95% CI 1.5-7.3; p<0.01), and restorative/advanced general dentists (OR 5.8, 95% CI 2.8-12.0; p<0.001) were associated with the number of information sources about DME (Table 3). Among these predictors, years of experience had the best predictive value but, with an R² of 0.3, can only be considered a relatively weak predictor.

Table 3. Effect of participant’s gender, education, training country, practice setting, years of experience, and specialty on the source of knowledge about DME and its predictors.

Knowledge about the materials used for DME

Eighty-seven percent of the cohort preferred to raise margins with adhesive restorations. In univariable analysis, all assessed parameters (educational attainment, training country, work setting, years of experience, and specialty) except gender were associated with knowledge about the material to use for DME (chi-squared test, p<0.05). However, in multivariable analysis, only educational attainment (OR 0.2, 95% CI 0.1-0.4; p<0.001) and 0 years of experience (OR 0.04, 95% CI 0.01-0.2; p<0.001) were associated with knowledge about the materials used to raise deep margins (Table 4). Years of experience had a medium effect size (R²=0.4) in predicting that this cohort of dentists would be inclined to choose adhesive restorations to raise deep margins.

Table 4. Effect of participant's gender, education, training country, practice setting, years of experience, and specialty on choice of DME restorative material and its predictors.