Introduction

In 2007, an international workgroup published new FDI criteria [1,2,3,4,5] to evaluate the quality of direct and indirect restorations; an update with clinical cases was published in 2010. This diagnostic system classified aesthetic, functional, and biological properties and covers various types of failures (Table 1) by using 16 different categories [4, 5] with five grades for each criterion. In detail, scores 1 to 3 indicated clinically acceptable restorations, and scores 4 and 5 summarized clinically unacceptable situations indicating repair (score 4) or replacement (score 5). The criteria were approved by the Science Committee of the FDI World Dental Federation (FDI) in 2007 and the General Assembly in 2008 as standard criteria that were specially designed for use in clinical studies [1,2,3,4,5]. The authors outlined the potential of the criteria to be applied 1) in evaluations of new restorative materials or operative techniques in clinical trials, 2) for quality assessment of dental restorations in daily dental practice (mainly in simplified form), and 3) under- and postgraduate education to determine whether a restoration needs reviewing, refurbishment, reseal, repair, or replacement [6] (Table 2). A recently published review [7] indicated a growing use of the FDI criteria in clinical trials, which increased from 4.5% in 2010 to 50.0% in 2016. In addition to this positive trend, it needs to be recognized that the criteria set was also assessed as complex with a lack of consistency in some parts [7] and several questions from users indicated the need for clarification. Aiming at increasing internal validity and promoting widespread dissemination for scientific, practical, and educational purposes, the expert group decided to review and revise the previously published FDI criteria set to improve the clinical usability, practicability, and acceptability. Beside the clarification of ambiguous issues, it was aimed to specify the recommendations for its interpretation and reporting.

Table 1 Common material-dependent failures
Table 2 Terminology and definitions and description of commonly used terms

Materials and methods

The existing FDI criteria [4, 5] have been improved by using a structured process to obtain information from a group of experts by means of a series of meetings and/or evaluations. This process with multiple rounds of feedback, open discussion, and rephrasing was iteratively continued until no further changes in the documents were needed [8]. A group consensus process is crucial in building guidance recommendations [9, 10]. In detail, the present information flow included a systematic search of the literature, a kick-off workshop under the participation of all experts as well as a structured communication flow aiming to converge existing opinions, and, finally, to reach a unanimous group consensus about the revised clinical criteria for the evaluation of direct and indirect dental restorations.

Expert panel

Ten experts in conservative and restorative dentistry agreed to participate, discuss, revise, and rephrase the criteria in spring 2019. Three of those (RH, SH, and AP) were also part of the original expert team. As several colleagues from previous projects [4, 5] were not available anymore due to different reasons, the work group was re-formed aiming at including experts from restorative dentistry of different regions.

J. Kühnisch and S. Mesinger coordinated the Delphi method as facilitators and collected all responses from the experts from the beginning, analyzed the opinions, structured the information, identified conflicting viewpoints; furthermore, they revised all documents accordingly. R. Hickel acted as a moderator during the workshops, online meetings, and discussions; furthermore, he provided numerous questions, comments, and suggestions by scientists, which he collected after the initial publications. Participants were forced to freely and consistently express their opinions and were encouraged to provide criticism or feedback and to detect errors or conflicting viewpoints. Although a consensus process captures collective knowledge, it should be noted that such criteria set may be, to some degree, a subjective viewpoint of the expert group [11].

Delphi method

The Delphi process was initiated with a group workshop at the Department of Conservative Dentistry and Periodontology in Munich, Germany, on June 3–4, 2019. During this face-to-face meeting, the existing scientific literature was presented and critically discussed, and empirical experiences of the existing scoring criteria for direct and indirect dental restorations were reviewed. In addition, a preliminary draft of a revised FDI criteria set was proposed based on the latest version [4, 5]. As a result of the workshop, the need and methodology for improvement were justified and agreed upon. After the meeting, the initially revised FDI criteria set was distributed, evaluated, and consistently updated. The following group discussion was held during an online meeting on September 16, 2019. The main intention of this meeting was to agree on the simplified structure and the importance of each category. This process continued until spring 2020, and the resulting criteria set was then pre-tested by the expert panel in a reproducibility study using intraoral photographs of different restorations with a broad spectrum of deficiencies. This study was performed in two rounds from May to July 2020. Feedback from the experts and statistical analyses of the intra- and inter-examiner reproducibility were compiled and discussed during other online meetings (July 21, 2020 and September 21, 2020). Further, where some inconsistencies or ambiguities were remarked, minor modifications were made to the FDI criteria set to harmonize the scores in each category. Diagnostic evaluations were repeated in a third round using the above set of clinical images. The final version of the revised FDI criteria was reviewed again by the whole expert panel and unanimously agreed on during another web meeting on November 9, 2020. The results of the reliability study were summarized in a separate report [12].

General considerations for clinical studies on dental restorations

In restorative dentistry, it is mainly evaluated how the material or restoration responds to the oral cavity of the patient with factors that may influence the success of the restoration, such as chewing forces, bruxism, diet, saliva, and the oral biofilm. Therefore, there are many confounders like patient factors, e.g., age, gender, tooth substance, chewing forces, oral hygiene, chewing tobacco, diet, general diseases, and local biological factors, e.g., location in the mouth, caries risk and periodontitis risk, and operator factors, e.g., clinical experience, decision making, and skills, which all influence the clinical performance of a dental restoration.

Study type and design

Several study types require a quality assessment of dental restorations. Here, clinical studies on new materials have to be mentioned primarily, which typically need a comprehensive restoration assessment after placement, during follow-ups, and at the final examination visit. Three- and 5-year follow-ups are at least advised for direct and indirect restorations, respectively. Longer observation periods are recommended especially when a new type of treatment or material is to be evaluated. For an observation period of 3 years, up to five recall sessions might be helpful. Ideally, the baseline evaluation should be carried out approximately 1 week after the insertion of the restoration and not during the placement appointment. If this procedure is not possible, the assessment by different dentists in the same appointment and an audio call interview 1 week later would be an acceptable compromise. Aiming to increase trial efficiency, baseline evaluation might also be performed after tooth rehydration approximately 30–60 min postoperatively and by checking the functionality no later than 4 weeks. The remaining recalls can be scheduled after (6), 12, 24, and 36 months. For longer observation periods, (bi)annual recalls might be preferable.

Furthermore, the quality of restorations could be evaluated in practice-based, epidemiological, observational, or diagnostic studies. In daily practice routines, practitioners are consistently forced to evaluate different aspects of restored teeth, which should be done with a validated and widely accepted set of criteria. When considering the whole spectrum of study types, it is understandable that the choice of categories and grades depends on each study’s intended purpose and methodological requirements. For clinical trials, the preferable examination setting is a dental unit with compressed air and standard illumination. Additional magnification tools, e.g., magnifying loupes, or documentation methods, e.g., intraoral photographs or 3D scans, may accompany visual examination. For practice-based studies, a simplified methodology might be more relevant. However, reporting of all chosen procedures is essential to better compare studies and interpret the results adequately.

Study population

It is recommended that clinical studies be conducted on the intended target population according to predefined and rigorously applied patient- and tooth-based inclusion and exclusion criteria, e.g., age range, gender, ethnicity, caries experience/risk or activity (high vs. low), parafunction or bruxism (present or not present), temporomandibular disorders (TMDs), oral hygiene (good, moderate, bad), smoking/vaping habits (no, moderate, heavy), or diet habits, e.g., coffee, tea, soft drinks, acidic foods, and beverages. Other habits of patients, such as frequent use of chewing tobacco or bubble gum, or parafunctions such as nail and/or thumb chewing, may also potentially influence the longevity of restorations and therefore need to be reported and re-evaluated with respect to the inclusion and exclusion criteria on each follow-up examination.

In addition to patient-related factors, it is essential to consider tooth-related variables. Here, the type of dentition (primary, mixed, permanent), tooth type (anterior, premolar, molar), quadrant, and affected surfaces are relevant. Furthermore, Black’s cavity class, the location of the cavity margin in relation to the gingiva (supra-, equi-, subgingival), and the hard tissues involved (enamel vs. dentin), the caries excavation technique and endpoint (selective vs. complete caries removal), as well as the type of antagonist teeth (unrestored vs. restored tooth, restoration material, not present) may be clinically relevant. Importantly, the indication to (re)place a restoration should be justified strictly according to common dental pathologies: 1) primary caries (proximal, occlusal, cervical, root, early childhood caries), 2) non-carious hard tissue defects, e.g., erosive tooth wear, abrasion, fractures/cracks or trauma, 3) dental developmental disorders, e.g., molar-incisor-hypomineralization or hereditary disorders of enamel/dentin, and/or 4) other specific situations, e.g., restorations to improve aesthetics due to discoloration or diastemas. The pooling of restorations with different characteristics in one clinical study, e.g., classes I and II, anterior and posterior teeth, or carious indications, e.g., caries and developmental disorders, should no longer be an accepted procedure. The flow of screened, eligible, and finally recruited patients/restorations should be described and illustrated as a flow chart according to the relevant reporting guideline for each study type, e.g., the CONSORT statement for randomized controlled trials [13, 14]. Beside this, patient’s motivation to adhere to the study protocol should to be safeguarded. Here, information cards might be helpful to provide data for the patient and dental professionals.

Evaluation of dental restorations

The quality assessment of dental restorations is a stepwise decision-making process that includes, if needed, the following procedures: 1) professional tooth cleaning and short air drying of the restored tooth for a few seconds, 2) functionality checks with standardized probes and blades, 3) static and dynamic occlusion testing with articulation paper, and 4) cold stimulus aiming at assessing hypersensitivity and pulpal reactions. It is also important to understand that the number of included categories can be chosen flexible according to the study aim and design. Furthermore, it can be decided if the scoring for each category will consist of five grades (excellent/good/satisfactory/unsatisfactory/poor) or, in a simplified form, only three grades (sufficient/acceptable = score 1 to 3, insufficient/inacceptable but repair possible = score 4, and insufficient/inacceptable but repair not possible/reasonable = score 5). The latter approach might be of relevance especially in practice-based studies. As some of the earlier described 16 categories were rarely used in clinical trials [7], the revision includes only the most frequently used ones now. The categories for general health, gingival, periodontal, and mucosal conditions, erosive tooth wear, or abrasion [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35] were separated from the “core” categories, as most of them are not directly related to the evaluation of dental restorations but reflect the status of the tissues beneath restored teeth (Table 3).

Table 3 Additional clinical parameters and the corresponding indices that might also be scored beneath dental restorations

Training and calibration

Clinical assessments in studies should always be carried out by trained and calibrated examiners. Therefore, appropriate theoretical and practical training sessions are mandatory and guarantee the consistency of judgments throughout the whole study period. Furthermore, the documentation of training is crucial. Each trainee should have a similar reproducibility rate in comparison to the trainer, which can be statistically expressed as intra- and inter-examiner reproducibility [36]. Clinical examples were published to assist study groups with this exercise [4, 5] and the revised FDI criteria set can be downloaded as illustrated document from the journal website. Nevertheless, calibration on patients in a clinical setting cannot be replaced by the evaluation of photographs, but time-consuming clinical calibration sessions might be shortened.

Recommended statistics

Studies on restoration quality and longevity require observations over a time, where different events, e.g., loss of patients, loss of teeth due to (non)study-related reasons, or failure of test restorations can occur. This implies an appropriate follow-up process and documentation of subjects, restorations, and failures. It is suggested to provide absolute numbers of failures and the overall number of evaluated restorations for each examination time point. There are different ways to calculate the mean annual failure rate or the normalized failure index besides the simple one dividing the total failure rate by the number of observation years [e.g., 37, 38]:

$$mean\;annual\;failure\;rate\;\left(mAFR\right)^1=1-\sqrt[t]{1-(\frac{N_{\mathrm{Failures}}}{N_{\mathrm{Restorations}}})}$$
$$mean\;annual\;failure\;rate\;\left(mAFR\right)^2=-\log(1-(\frac{N_{\mathrm{Failures}}}{N_{\mathrm{Restorations}}}))/\mathrm t$$
$$Normalized\;Failure\;Index\;(NFI)=\frac{N_{\mathrm{Failures}}}{(N_{\mathrm{Restorations}}\ast t)}$$
N Failures :

total number of failed restorations

N Restorations :

total number of investigated restorations

t :

observation time

1 :

preference of this formula in case of low failure rates

2 :

preference of this formula in case of high (almost 100%) failure rates in less than 1 year.

For calculating of the success rate, the dichotomization of the data into sufficient (scores 1–3) and insufficient (scores 4 and 5) is needed. The calculation of the survival rate uses the dichotomization of the data into restoration present including repaired (scores 1–4) and not present/failed (score 5). Kaplan–Meier curves are frequently applied to illustrate the success or survival probability over time [36]. The log-rank test is usable to compare differences between groups [39,40,41]. In addition, Bonferroni corrections or multivariate analyses, e.g., Cox proportional hazards model or Poisson distribution can be computed. When considering the potential influence of all patient-related factors on restoration survival, it is recommended to conduct a multiple logistic regression analysis.

Following the aim of increasing the internal validity domains and categories is somewhat rearranged in relation to their clinical relevance and importance; therefore, the functional properties (domain F) were now placed at the beginning of the assessment followed by the biological (domain B) and the aesthetic properties (domain A). The revised FDI core criteria set summarizes 11 criteria. In addition, the criteria “patient’s view” and “radiographic evaluation” were shifted in the new domain “miscellaneous” (domain M).

Domain F: functional properties

The assessment of the function of a restoration is a key issue in scientific studies as well as in daily dental practice. Here, the visual examination provides relevant information that is sometimes hard to objectify and quantify. Therefore, the use of metric instruments improves the validity of the criteria. For this purpose, standardized instruments, e.g., metric probes and blades, are recommended to use the FDI criteria reliably.

Fracture of material and retention (category F1)

Restoration fracture and retention are the most relevant categories in clinical practice when evaluating direct and indirect restorations and therefore should be included in any study. Different fracture patterns and retention failures may occur in relation to the type of restoration: cracks, chipping/delamination, bulk fractures, or incomplete and complete loss of retention (Table 1). Minor material chipping or hairline cracks, which sometimes can only be detected after tooth cleaning and air drying, most often do not require an operational intervention, but these events should be monitored in follow-up visits recorded during data capture. Small chipping fractures with loss of material might only be monitored or corrected by refurbishment, e.g., recontouring and polishing. The main reason for failures of direct composite restorations is bulk fractures [41, 42], which can potentially be repaired. In cases of severe or multiple bulk fractures, replacement of direct restorations is considered the treatment of choice [41,42,43,44]. Types of fracture patterns are sometimes different in indirect restorations (Table 1). Material chipping of variable extension is quite common in veneered ceramic restoration. In monolithic ceramic restorations, bulk fractures are more common. Ceramic fixed partial dentures primarily fracture in the connector area [45, 46]. Bulk fractures or delamination of a greater volume may substantially affect restoration integrity. If the loss of material is localized, repair might be possible. Repair of a restoration with extended or multiple fractures might not be reasonable, and complete replacement is more appropriate.

Severe loss of retention is in any case insufficient (scores 4 and 5), but the extent of lost material, partially or (almost) completely, defines whether a direct restoration might be repaired or not. If a restoration is graded as completely loose or lost (score 5), all other functional and aesthetic categories usually become not applicable. Indirect restorations, which can be recemented/reluted, will be rated with score 4 (repair).

Marginal adaptation (category F2)

There are different interfaces between the dental hard tissue, restorative material, and adhesive and/or luting resin/cement layer. Each interface can degrade and potentially alter marginal adaptation. In clinical practice, it is impossible to distinguish failures between the different interfaces. Therefore, only the marginal adaptation as such can be assessed. The quality of marginal adaptation is both the result of the properties of the adhesive, luting resin/cement, and restorative material and the skill and knowledge of the operator to create a good restoration (adequate cavity preparation, moisture control, application of materials according to instructions for use) [1,2,3,4,5, 41].

Evaluation of marginal adaptation should be done by visual examination and the use of a metric 250-µm probe, e.g., Fissuren Sonde 250EX with 250 µm diameter (Deppeler, Rolle, Switzerland). With respect to practicability, another probe with a diameter of 150 µm is no longer preferred. Ideal marginal adaptation shows a smooth transition from the restoration material to the surrounding tooth structure; no marginal irregularities should be detectable by gentle probing. Minor marginal deficiencies can be detected as discoloured margins or ditches and will be categorized as “sufficient” [47]. Wide (> 250 µm) marginal gaps with a gap depth ≥ 2 mm indicate a situation of clinical insufficiency and probably require dental intervention depending on both the location and the caries risk/activity/history of the patient [4, 5, 48,49,50].

Proximal contact point (category F3)

The tightness of proximal contact points should be estimated in a reproducible manner. Metal blades (e.g., matrix for EX kit; blades’ thickness 0.025 0.05, and 0.1 mm; Deppeler, Rolle, Switzerland) are recommended for better categorization [1,2,3,4,5, 51]. In case of unavailability of the blades, waxed dental floss might be considered a non-standardized alternative. A proximal contact point has a physiological strength when the 25-μm metal blade (or dental floss) can pass through it with resistance [1,2,3,4,5]. An appropriate degree of contact strength as well as a properly located contact area is recommended to prevent food impaction and allow for interdental papilla to fill the interproximal space [52, 53]. The lack of physiological contact point strength may result in food impaction, papillitis, or discomfort. However, it must be addressed that physiological contact strength can vary considerably among patients [54]; therefore, the strength of proximal contact points should be assessed individually and with caution. For better evaluation, an adjacent contact for comparison could be used. In addition to less optimally restored contact areas, weak or no contact points could be linked to the individual tooth form, e.g., microdens, atypical tooth position, diastema, and/or paced/gap-toothed dentition. In these clinical situations, this criterion shall not be applied. The same applies to patients with advanced periodontitis or mobile, flared, or missing teeth. Generally, teeth with nonexisting proximal contacts cannot be evaluated with regard to proximal contact points.

An unintentionally interlocked contact point due to excessive restorative material, bonding agent, luting resin, or cement, which makes it impossible for a blade or dental floss to pass, has been added to the revised FDI criteria set. Unintentionally interlocked contact points are unacceptable, as they impede oral hygiene, make affected tooth surfaces inaccessible for proper cleaning, and may therefore cause caries and/or periodontitis.

Form and contour (category F4)

In modification to the previously published recommendations [1,2,3,4,5], this category is now listed under “functional properties,” because both are essential variables of the physiological functionality of any restoration in the masticatory system [55]. As characteristics of an optimally restored tooth form and contour, the following indicators of functionality have to be mentioned: (1) gingiva and periodontium are protected, (2) physiological embrasures between teeth are rebuilt and potentially allow for an alignment of the interdental papillae [52], (3) a spillway for the passage of food during mastication is reconstructed and prevents gingival food impaction, (4) the restoration safeguards the self-cleansing ability and the occlusal embrasure allows for better access for oral hygiene floss passage, and (5) stabilizes the position of the tooth to adjacent and antagonistic teeth. It needs to be further noted that optimal reconstruction of form and contour not only guarantees functionality but also substantially affects aesthetics. In anterior teeth, angulation and width-to-height ratio should be additionally considered [56].

The individual rebuilding of form and contour depends on the patients’ wishes as well as on the dentist’s or dental technician’s skills. An ideal form and contour may not be achievable in the case of children/adolescents or elderly individuals with reduced compliance, disabled patients, individuals with dental anxiety, or patients with limited mouth opening. Furthermore, irregular tooth angulation and/or position in the jaw, e.g., due to tooth crowding, may also complicate an ideal restoration in terms of form and contour. Irregularities, including overhangs and positive steps of the restoration, should be improved by refurbishing to avoid negative side effects, e.g., plaque accumulation and marginal discoloration. Underfilling might be the result of a primarily under-contoured restoration or/and of a gradual process of deterioration of the restorative material.

Occlusion and wear (category F5)

The static and dynamic occlusion of a restored tooth influences the functionality of the dentition. An ideal occlusion of the restoration should be harmonized with the individual and age-related occlusion of the masticatory system. The restoration with antagonistic teeth should not have a non- or hyper-occlusion; it should avoid biomechanical stress on the supportive tissues and should not trigger pain and TMDs. A restoration with non-occlusion potentially limits the chewing ability and may result in the elongation/super eruption of the restored and/or the antagonist tooth if there is no contact to prevent this. Oversized cusps or dimensions of the occlusal surface could lead to premature contacts, hyper-occlusion, and interfering balances, which may negatively influence the restoration’s longevity in terms of material chips or fractures (category F1) and may cause discomfort, pain, or TMDs [57,58,59,60,61,62,63].

Wear is the result of dynamic processes predominantly on occlusal and proximal surfaces of the restoration, which might be influenced by individual occlusion, bruxism, individual habits, and nutritional/chemical and mechanical challenges. Wear can hardly be assessed by clinical examination alone; therefore, objective monitoring methods are required, which can directly compare follow-up with baseline information, e.g., plaster models or 3D scans [64]. Therefore, only a simplified recording in the revised FDI criteria set has been integrated. If quantitative information on wear is needed, intraoral 3D scans or scans of replicas after impression-taking should be considered as the method of choice [65,66,67].

Domain B: biological properties

Pathological processes that are related to dental restorations include caries at the restoration margins (B1), dental hard tissue defects, cracks or fractures (B2), and postoperative hypersensitivity or pulpal inflammation (B3). Several other dental pathologies, e.g., developmental dental defects, bruxism, and erosive tooth wear, may potentially interfere with longevity. With respect to the aim of clarification and prioritization, the most frequent pathologies are covered in the revised criteria set. Therefore, particular research questions may require the inclusion of additional standard methods (Table 3).

Caries at restoration margins (CAR, category B1)

This category has been harmonized in relation to the current caries definitions [28, 68,69,70,71]. Caries is the most prevalent dental disease from a global perspective [72,73,74], and the etiology of caries at restoration margins (CAR, synonyms: secondary caries or recurrent caries) is not different from that of primary caries [75,76,77,78]. CAR is mostly located in plaque stagnation niches, e.g., proximal margins, and can rarely be diagnosed on smooth surfaces that are well accessible for oral hygiene [79]. Furthermore, it may require significant gaps that are accessible to oral fluids in a caries active oral cavity to contribute to the risk for CAR. Some early clinical studies conclude to minimum gap width occlusally of 400 microns and 250 µm approximally [1, 2, 48], but recent in situ studies also show that gaps as small as 50 microns might be able to generate CAR [80, 81]. For margins at the proximal gingival box of a class II restoration, smaller gaps or even no gap maybe associated with caries at this site. The most important factor regarding CAR is the caries activity of the patient.

Clinically, CAR reaches from non-cavitated carious lesions to deep cavities. While initial carious lesions require no (reviewing, topical fluoride application) or minimally invasive intervention reseal or refurbishment, cavitated lesions at the restoration margin probably need operative dental measures in terms of repair or replacement depending on the size of the defect and restoration and caries activity of the patient [71, 82, 83]. Importantly, if any caries occurs at any other site of a tooth that is not directly related to the restoration, it should not be registered as CAR. The clinical diagnosis of CAR is sometimes difficult to differentiate from stained margins [76, 77]. Importantly, stained restoration margins with no demineralized hard tissue should not be confused with CAR.

Dental hard tissue defects at the restoration margin (category B2)

This criterion comprises tooth cracks, enamel chipping, or cusp fractures at the restoration margin. Additionally, cracked tooth syndrome [84] is considered, which may also cause hypersensitivities or pain. In addition, it is noteworthy not to include other events in this category, such as physiological attrition and wear, abfraction, or defects related to other reasons, e.g., trauma. Additionally, a lost restoration material or CAR must be scored in their corresponding categories. Clinical assessment could be supported by light transillumination of the restored tooth.

Postoperative hypersensitivity and pulpal status (category B3)

Postoperative hypersensitivity is linked to pulpal reactions immediately after placement of a dental restoration and can include discomfort, pain, pulpitis, or, later, loss of tooth vitality. There are several factors that affect the pulp-dentin complex: 1) diagnosis and history of the tooth, 2) dental treatment including cavity preparation, caries excavation, or placement of a properly sealed restoration, 3) properties of the adhesive, luting resin, and/or restoration material, and 4) the patient’s individual pain perception. In general, the diagnosis of postoperative hypersensitivity may indicate the presence of a deficiency during the restoration workflow, e.g., incomplete adhesive bonding of the restoration, which is probably difficult to identify later.

With respect to definition, tooth sensitivity needs to be recorded before and after restoration placement and at all recall visits. On each examination, it is necessary to consider, first, the patient’s reporting of tooth (hyper)sensitivity, e.g., by using a visual analogue scale, and second, testing of irritability of the pulpal nerve on cold, e.g., with dry ice or cold spray, in comparison to the reaction of a contralateral, sound, and unrestored tooth. The restoration should be rated as acceptable when normal sensitivity or mild pulpal symptoms are recorded during follow-up examination. In cases of postoperative hypersensitivities, transient pain or more intense pulpal reaction, individual monitoring intervals might be indicated. Irreversible pulpitis or pulp necrosis requires endodontic intervention to overcome the problem.

Domain A: aesthetic properties

The aesthetic performance of dental restorations can be characterized by surface luster, surface texture, marginal staining, color match, and anatomical form. The evaluation is somewhat subjective and therefore more prone to potential bias and variability [85, 86]. The aesthetic appearance of a restoration depends mainly on how well it blends into the surrounding tooth structure, which is influenced by oral hygiene.

The evaluation of aesthetic properties is of clinical relevance for visible and tooth-colored restorations within the smile frame only, usually canine to canine. In many patients, the mesiobuccal aspect of upper premolars is visible when patients smile and therefore essential for aesthetic appearance. In most individuals, however, the evaluation of aesthetics in posterior teeth is less important. Depending on the study design, setting, and aim of the investigation, researchers can choose if the evaluation of the aesthetic properties should be evaluated from a standard examination distance under operating light (~ 40 cm) or from a speaking distance (~ 80–100 cm), which will lead to different results and should therefore be mentioned. Additional devices to objectify aesthetics are intraoral photographs or scans, color scales, colorimeters, spectrophotometers, or 3D imaging.

Surface luster and surface texture (category A1)

Surface luster and texture are created by the reflection of light from the surface of the restoration, which mainly depends on material properties and the restoration surface [87]. In detail, the intrinsic material roughness (nano- and micro-roughness), finishing procedures (macro-roughness), e.g., polishing marks, and flaws due to material properties or material processing, e.g., pores and voids, must be considered. Macroscopic deviations in surface texture, such as polishing marks or pores, are easier to detect by visual examination than minor deficiencies [87, 88]. Ideally, the surface luster and texture of the restoration are comparable to that of the surrounding hard tissue.

Marginal staining (category A2)

Marginal staining is defined as the discoloration of a crevice between the cavity wall and the restoration, subsequently affecting the margin of the restoration, which should not be confused with caries [76, 77]. A prerequisite for staining is the presence of a ditch or gap at the margins where pigments can adhere. Marginal staining depends on the efficacy of the adhesive/cementation system to bond the restoration to dental hard tissue(s) and individual patient factors [41]. The latter include nutritional habits such as consumption of coffee, black tea, or red wine as well as smoking and oral hygiene procedures [89, 90]. Additionally, the individual intraoral microbiome may play a role [91,92,93]. Less important is the restorative material [94,95,96] or the chosen operative technique [97, 98]. Nevertheless, there is evidence that suggests that the occurrence of marginal discolouration correlates with a compromised integrity of the marginal seal [47, 99], which may be frequently related to polymerization shrinkage of the composite.

Colour match (category A3)

This category is applicable to tooth-colored restorations only. An ideal color match is achieved when all visually apparent differences between dental hard tissues and the restorative material are minimal or even invisible. Deviations in shade, translucency, or opacity between dental hard tissues and the restorative material are possible if (1) the chosen color of the restorative material does not match that of the surrounding dental hard tissues, (2) the natural teeth become darker or more yellow with increasing age [100], and (3) the restorative material itself has inherent color instability [101,102,103,104]. When color matching has to be evaluated, visual examination is the method of choice. In addition, intraoral photographs can be used but are also difficult to standardize during follow-up examinations [105, 106]. In contrast, commercially available color measuring instruments, e.g., reflectance spectrophotometers and colorimeters, have gained acceptance due to their satisfactory accuracy, reliability, and time-efficient use [86, 107,108,109].

Domain M: miscellaneous

The expert panel decided to streamline the “core” FDI criteria set and additional methods are listed in Table 3. The patient’s view on the restored tooth as well as the radiographic assessment of restorations was shifted in a new domain. With respect to the impossibility to embed the corresponding diagnostic scores into the standard 5-point scale, both categories are shown in the illustrated version only, which can be downloaded from the journal’s website.

Patient’s view (category M1)

Patient satisfaction with a dental restoration is a subjective response that gains more attention in practice-based or health service research and is usually scored by means of visual analogue scales [e.g., 113]. From the methodological point of view, it might be sufficient to ask for an overall (subjective) impression from the patient. In cases of dissatisfaction, a detailed report about pain, hypersensitivity, chewing comfort, occlusion, proximal contacts, cleanability, contours, or aesthetics might be of value. This assessment can be designed by use of the FDI criteria but a standardized protocol is not established or published so far. The patient’s opinion might be relevant, especially if the aesthetics of the restorations appear to be unacceptable for him/her and a replacement ahead of time needs to be discussed. It should be noted that the patient’s view can interfere with dental assessment and clinical decision-making.

Assessment of dental restoration on radiographs (category M2)

In general, it has to be emphasized that there is no general justification to do a radiographic examination for the assessment of dental restorations without any clinical indication [111,112,113,114]. This approach is strictly in line with basic principles of radiation protection [115,116,117]. Nevertheless, the assessment of direct and indirection restorations is required on justified images. Here, radiographic evaluation includes, among others, caries detection, negative/positive steps or marginal gaps of the restoration, apical periodontitis, periodontal bone loss, internal/external resorption, or quality of endodontic treatment.

Interpretation of the scorings

In addition to the intention to objectify the diagnostic evaluation and assessment of dental restorations and assist clinicians in decision-making, it is also important to consider clinically relevant key information, e.g., caries risk and activity, age, and medical or behavioral problems. On the basis of this comprehensive information, an individual intervention strategy has to be justified and agreed upon between the dentist and the patient/caregivers knowing well that the final decision might also be influenced by varying diagnoses, treatment philosophies, experiences, settings, and available resources including treatment costs. Importantly, each decision must be made with respect for the patient’s autonomy. Therefore, a specific dental diagnosis might be linked with different decisions.

With the revised FDI criteria set (Table 4), some ambiguities were removed, and scores were further harmonized to cross-link distinct clinical situations with possible management strategies, e.g., monitoring/reviewing (scores 1–4), refurbishment or reseal (score 3), repair (score 4), or replacement (score 5). An important issue is the decision whether a restoration is clinically acceptable (scores 1–3) or not (scores 4 and 5) and to decide further whether repair is possible (score 4) or not (score 5). Again, as described above, treatment procedures have to be understood as possible intervention corridors, and they are not meant to be understood as inevitable treatment approaches. In that respect it may be also good to consider contemporary tendencies in restorative dentistry to be as conservative/preservative as possible and, in case of doubt, rather select shorter monitoring intervals or the most minimally invasive option [118]. In this context, repair of direct and indirect restorations has to be considered a conservative treatment option in comparison to traditional replacement.

Table 4 Description and details of the revised FDI criteria set to evaluate direct and indirect dental restorations

Reporting of future studies

In addition to the detailed description of the FDI criteria set and the clinical interpretation of the diagnostic findings, it is vital to highlight the need for standardized study reporting that includes the evaluation of dental restorations. A reporting checklist is given in Table 5, which should help researchers to standardize their paper writing.

Table 5 Reporting checklist for studies evaluating direct and indirect restorations

Conclusions

The formerly published FDI criteria set for the evaluation of direct and indirect restorations [1,2,3,4,5] was revisited through a stepwise consensus process. With the aim of improving clinical usability, practicability, and acceptability, a revised set of criteria prioritized categories and harmonized the wording. It is also important that each domain or category can be selected independently, thus creating a modular diagnostic system with great flexibility for the evaluation of direct and indirect restorations. The revised FDI criteria set has to be understood as a living document that can be regularly adopted on the basis of new clinical data, findings and experiences. Therefore, we encourage researchers, teachers, and dental practitioners to provide feedback.