How to Intervene in the Caries Process: Early Childhood Caries – A Systematic Review

Early childhood caries (ECC) is a specific form of caries that occurs in children on primary teeth. ECC is worldwide and a highly prevalent disease [Chen et al., 2019] and should be considered a relevant burden for the children and their families [Thomson, 2016]. ECC is characterized as “the early onset of caries in young children with often fast progression which can finally result in complete destruction of the primary dentition” [Machiulskiene et al., 2019]. The aetiology of ECC is quite clear: high sugar intake usually via a nursing bottle and insufficient or no oral hygiene, as parents do not brush or not well enough, leading to an atypical pattern of caries attack, particularly on smooth surfaces of upper anterior teeth in young children [Drury et al., 1999; Tinanoff et al., 2019; Wyne, 1999; Machiulskiene et al., 2019]. In contrast to the ECC classification regarding typical clinical patterns [Wyne, 1999], the epidemiological ECC classification of ECC records any carious lesions in primary teeth in children younger than 6 years [AAPD classification]. Many studies indicate that severe ECC can negatively influence the quality of life in preschool children, but fortunately be improved when treated [Collado et al., 2017]. Moreso, due to young age and often low cooperation, general anaesthesia (GA) or sedation is frequently necessary for invasive dental interventions (e.g., fillings, pulp therapy, extraction), going along with higher risks for the children and efforts for families, and also considerable costs for the health systems [Casamassimo et al., 2009; Thomson, 2016]. This leads to a more complex dental decision making in how to manage ECC as not only the diagnosis of the caries lesion at the tooth level but also the caries management on the patient level needs to be considered. Furthermore, dentists face a lot of challenges when planning operative interventions of (severe) ECC:

• The morphology of primary teeth in comparison with permanent teeth is different; as they exhibit large pulp chamber and a thinner enamel–dentin layer, which shorten the time for the caries process to reach the pulp and may cause pain and pulp complications.

• Pain and fear caused by ECC in emergency dental care can compromise future oral health-related quality of life of the child.

• Young children are immature, and usually pre-cooperative (sedation or GA might be needed), and a more complex decision-making process as parents are responsible to take care for their child.

• Premature loss of anterior primary teeth may cause problems with eating/biting, difficulties in speech development, and retarded eruption of permanent incisors.

Another paper in this review consortium for a consensus paper of the European Organisation for Caries Research and the European Federation of Conservative Dentistry deals especially with the management of carious primary molars and different methods of removal of carious tissue and the subsequent restorations. Therefore, this topic is not included in this systematic review. This review focuses on the management of ECC, especially carious primary (upper) anterior teeth.

The aim of this review was to systematically evaluate the current state of knowledge on how to intervene in the caries process in carious primary (upper anterior) teeth in ECC.

This systematic review aims to answer the following Participants, Interventions, Comparisons, and Outcomes questions:

How to intervene in the caries process in ECC (especially upper anterior teeth)?:

1. What is the effect of non-invasive management options on ECC?

2. What are the success rates of restorative management options of ECC?

The authors discussed the review protocol a priori. No further external registration was performed as the topic was given to the authors by the joined chairs of the Organisation for Caries Research/European Federation of Conservative Dentistry Consensus Workshop on how to intervene in the caries process.

For the comprehensive search strategy, Medline was searched via PubMed for appropriate papers up to and including March 2019. The used search terms and the full search strategy are shown in Figure 1. Titles and abstracts were screened to exclude papers not related to the topic. The remaining full-text articles were screened for eligibility and references hand-searched for additional sources. Publications on humans in English language were included.

The inclusion criteria were (Participants, Interventions, Comparisons, and Outcomes):

• Patients: preschool children, ECC, carious incisors, non-cavitated and cavitated (anterior) primary teeth

• Intervention:

• Nonoperative management of carious lesions in primary (anterior) teeth with silver diamine fluoride (SDF) or any other nonoperative strategies or measures of secondary prevention;

• Restorative management of carious lesions in anterior teeth.

• Control:

• Any other nonoperative strategy (or if available restorative approach [RA]);

• Any other restorative material.

• Outcome:

• Caries or lesion progression (major failure, when necessity for pulpal therapy);

• Survival of restoration (major failure, when pain/necessity for pulpal therapy).

• Meta-analyses, systematic reviews, and in case of no studies for these high evidence levels also randomized controlled trials (RCTs) and/or cohort studies (prospective and retrospective).

The exclusion criteria were:

• Permanent teeth, schoolchildren, adults;

• ECC primary prevention, (vast majority of participants with) caries-free primary teeth at baseline;

• Only caries lesions in primary molars;

• Case presentations, case series;

• No clinical outcomes reported.

The title and abstract lists, containing 401 hits, were independently assessed by 2 authors (K.G. and J.S.), respectively. Each of them screened the titles, abstracts, full-texts of eligible papers regarding management of ECC categorized into subchapter (1) nonoperative management (SDF and other nonoperative caries management [NOCM]) and (2) RAs. Papers with questionable inclusion were discussed in telephone conferences among the authors (K.G., J.S., C.H.S., and H.J.).

Eligible papers were read in full-text by the authors, especially for methodology, bias, and outcome. Disagreement between the authors was resolved by discussion in additional telephone conferences. The papers that fulfilled all of the selection criteria were processed for data abstraction (Fig. 1).

Information extracted from the studies included publication details, setting, participants, focused question, search results, comparators, main outcomes, and conclusions (Tables 1-3).

Depending on the type of studies, the reviewers (K.G. and J.S.) assessed and classified the risk of bias, mostly according to the PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health-care intervention [Liberati et al., 2009] by scoring the reporting and methodological quality of the included systematic reviews on NOCM with SDF (online suppl. Table 1; for all online suppl. material, see www.karger.com/doi/10.1159/000504335). A list of 27 items was assessed, and if all individual items were given a positive rating by summing these items an overall score of 100% was obtained. Only systematic reviews including meta-analysis could achieve a full score of 100%. The estimated risk of bias was interpreted as follows: 0–40%: high risk of bias; 40–60%: substantial risk of bias; 60–80%: moderate risk of bias; 80–100%: low risk of bias.

Only few studies with higher level of evidence reporting outcomes on NOCM and operative options were found, to at least gain some impressions of the results of these various management options not only systematic reviews and RCTs were included. Due to the diversity of the study types, the quality assessment for the nonoperative and operative management of ECC was performed (C.H.S. and J.S.) according to the Newcastle-Ottawa Scale (online suppl. Tables 2, 3), which is primarily aiming at assessing the quality of nonrandomized studies [Wells et al., 2016].

The heterogeneity across studies was investigated according to the following factors as retrieved by the included literature:

• Study and subject characteristics;

• Methodological heterogeneity (variability in study design and risk of bias);

• Analysis performed (descriptive or meta-analysis).

Due to the availability of a higher level of evidence for SDF (systematic reviews/meta-analyses) and the fact that in contrast to other NOCM options the effect of SDF was well examined for cavitated carious lesions, this subchapter of NOCM was presented individually (Table 1), whereas other NOCM options are shown together (Table 2). For SDF and other measures of NOCM, the outcome was mainly caries arrest (e.g., hardness, colour, lesion size), and for RA, the main outcome was survival of restoration, that is, restorations not needing any restorative reintervention (replacement or repair).

The results are presented for SDF, NOCM, and RA separately, as different kinds of studies with different levels of evidence were found for the different aspects in managing ECC on the tooth level. No RCT investigated NOCM options versus RA.

The application of SDF for ECC has been investigated in many studies. Within the selected manuscripts, there were 5 systematic reviews (Table 1), of which, 3 [Oliveira et al., 2019; Chibinski et al., 2017; Gao et al., 2016] contained a meta-analysis. All of the reviews containing a total of 22 different RCTs which homogenously show the high potential for caries arrest of SDF. The higher the concentration of SDF, the better the arrest rates, especially when applied several times, for example, biannually. SDF solution applied once or twice a year could arrest active caries lesions involving dentine with a success rate of 79 and 91%, respectively [Duangthip et al., 2015]. Adverse effects such as black stains were reported in studies using 38% SDF [Duangthip et al., 2015].

Other NOCM approaches than SDF mostly employed fluoride varnish. For this subchapter, 10 studies were included (Table 2): one systematic review with a part of a meta-analysis [Gao et al., 2016], 5 RCTs, 2 cohort studies, and 2 control cohorts in SDF-RCTs [Duangthip et al., 2018; Chu et al., 2002]. The included studies showed that at least 63.6% [Gao et al., 2016] up to 81.2% [Autio-Gold and Courts, 2001] of enamel carious lesions were inactivated after fluoride varnish application. For cavitated lesions, fluoride varnish was effective in 30% of cases [Duangthip et al., 2018; Lo et al., 1998]. Sodium fluoride varnish has a moderate effect in remineralizing and/or arresting early enamel carious lesions (Table 2) [Gao et al., 2016; Autio-Gold and Courts, 2001; Sitthisettapong et al., 2012; Memarpour et al., 2015], and a lower effect for cavitated carious lesions [Divaris et al., 2013; Duangthip et al., 2018]. Moreso, by brushing with fluoride toothpaste (1,000 ppm), the arrest of ECC may be achieved [Lo et al., 1998; Peretz and Gluck, 2006], but the effect is low with a low level of evidence. There is also limited evidence for efficacy of casein phosphopeptide-amorphous calcium phosphate: Daily application of 10% casein phosphopeptide-amorphous calcium phosphate paste added to regular toothbrushing with a fluoride toothpaste showed no additional effect in arresting enamel caries in the primary anterior teeth [Sitthisettapong et al., 2012], contrary to the findings of [Memarpour et al., 2015].

Regarding restorative options in carious anterior upper primary teeth, 8 studies (5 RCTs, 3 cohort studies; Table 3) were included. The level of evidence was assessed as low with a high potential of bias. In many of the studies, restorations were performed under GA (Table 3). Cohort studies showed success rates of composite restorations on tooth level of 50–90% (81.5%, ∼2.5 years [Buecher et al., 2013; Ram and Fuks, 2006]). Looking into more details, multisurface composite restorations had clearly lower retention rates (<70% [Buecher et al., 2013], 80% [Ram and Fuks, 2006], 90.9% [Amin et al., 2016]) in anterior teeth than single surface resin composite restorations (>90% [Buecher et al., 2013; Amin et al., 2016]). Strip crowns had higher success rates (>80%, 2 years [Ram and Fuks, 2006], 78% [Walia et al., 2014]) than class III and IV fillings (67.8% [Buecher et al., 2013]), but lower then preveneered stainless steel and zirconia crowns (95%, 100%, respectively 1/2 year [Walia et al., 2014]). Lower retention rates were found for GIC fillings (17% failure, 1 year), regardless if atraumatic restorative treatment or complete removal of carious tissue under local anaesthesia was used [Arrow et al., 2016]. For the use of posts in restoring carious primary maxillary anterior teeth with composite strip crowns, 12 months success rates of 63–98% were reported [Eshghi et al., 2013; Sa-want et al., 2017].

The estimated risk of bias regarding systematic reviews on SDF is presented in online supplementary Table 1, showing moderate to low potential risk of bias and homogenous results.

Due to the diverse types of studies on NOCM (Table 2) and RAs (Table 3), the Newcastle-Ottawa Scale [Wells et al., 2016] was used for quality assessment (online suppl. Table 2, 3). Overall, considerable heterogeneity was observed in the NOCM and RAs and materials under analysis, the subject characteristics and outcome measures. Furthermore, quality assessment scales and reporting of effect scores varied among the publications. The results of NOCM were biologically and clinically plausible and mostly in line with each other, as no outlying heterogeneity could be assessed. Nonetheless, a potential for a high risk of bias is present. Regarding operative approaches in ECC, the results were plausible for clinicians and not contradictory, while a potential for a high risk of bias is present due to the study type and underlying confounding factors.

The level of evidence emerging from this systematic review indicates that there is high evidence to support a high effect of SDF in arresting, especially cavitated carious lesions, a low level of evidence for a moderate effect of other NOCM options of ECC, especially non-cavitated lesions, as well as a low level of evidence of moderately high and clinical relevant failure rates for restorative care.

ECC is highly prevalent worldwide and clearly a marker for high caries risks [Chen et al., 2019]. Therefore, it was not surprising that the literature search revealed a large number of studies, but the majority was excluded for this review as they dealt with the epidemiology or the primary prevention of ECC. On the other hand, it was surprising that the management options of ECC have not been investigated profoundly. In addition, rather rare options of posts in primary incisors have been examined in RCTs [Eshghi et al., 2013; Sawant et al., 2017], whereas frequently performed options such as fluoride varnish and fillings (glass ionomer cement, composite) have not been compared with each other in high-quality studies.

Management of cavitated carious lesions in primary teeth of very young children could be performed with SDF, as its effectiveness in arresting the activity of carious lesions (cavitated and non-cavitated) is clearly shown (Table 1). It is easy to apply, even outside dental offices, and relatively inexpensive solutions can be used, especially when compared to restorative intervention under GA. The few but relevant drawbacks of SDF are that: (1) not in all countries a SDF product is available or not to be used for caries management (off-label use only) and (2) the occurring black staining can result in a lower acceptability for caries management by parents/caretakers.

Nonoperative management of ECC has also shown to have moderate potential to inactivate the carious process, especially in enamel lesions (Table 2). The process of inactivation of dentinal carious lesions occurs clearly slower and less predictable, when compared with the application of SDF. Similarly to the findings in this review (Tables 1, 2), the American evidence-based clinical practice guideline for NOCM in primary teeth strongly recommends 38% SDF for any cavitated carious lesion and 5% sodium fluoride varnish for non-cavitated enamel lesions [Slayton et al., 2018].

The combination of daily toothbrushing with fluoride toothpaste at home and fluoride varnish application in office or kindergartens has the advantageous effect that healthy behaviours such as daily toothbrushing and fluoride use are trained in recall visits, which may influence conviction of parental self-efficacy, laying the responsibility of treatment of ECC into caregivers hands away from the dentist (SDF or restoration). Adherence to oral hygiene instructions and diet counselling are essential in caries management on patient level and should be addressed for long-term effects. Guiding the awareness and understanding of one’s responsibility are important for a potential long-term oral health in the children, making it clear that despite other management options NOCM should be obligatory for all children with ECC as an approach of secondary prevention on a child and tooth -level.

RAs for the management of ECC are possible, but there was low scientific evidence to suggest that conventional composite fillings for anterior teeth are a valid surgical option in ECC. There is also insufficient information about atraumatic restorative treatment in anterior primary teeth leaving little scientific evidence for the choice of restoration [Waggoner, 2015] or a restoration at all instead of an extraction. Only in the studies where a post was used followed by a composite strip crown, higher success rates could be achieved (Table 3).

The RA often masks the caries activity in the child with ECC as the restorations show a high risk of failure frequently due to secondary carious lesions. When taking moderate success rates of 50–90% on the tooth level even for fillings performed under optimal circumstances such as GA (Table 3), these rates diminish to very low success rates on a patient level due to the usually high numbers of affected teeth in one child. GICs that are less technique sensitive and being placed in only one increment could restore and arrest active dentin carious lesions, but there was low scientific evidence to suggest GIC as material of choice for anterior primary teeth [Zhi et al., 2012]. When a RA is chosen for multisurface cavities, composite strip crowns should be favoured over fillings, but possibly preveneered stainless steel crowns or prefabricated primary zirconia crowns have higher success rates [Walia et al., 2014]. For RAs due to very young age and deep carious lesions with potential pulpal involvement invasive treatment may frequently be performed under sedation or GA with all its risks, efforts and costs, which should always play a role in the individual decision-making process on how to intervene in the carious process of ECC.

Measures of secondary prevention should always be applied when managing ECC. Additionally, fluoride varnish is to be suggested for non-cavitated lesions and SDF for cavitated dentine lesions without irreversible pulpal involvement, while decisions for RAs should consider patient-related risk factors and the likelihood for the necessity of sedation or GA.

The authors especially thank David Manton for contribution in the search string.

This research complies with the guidelines for human studies and was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

Authors declare no conflicts of interest.

J.S., K.G., C.H.S., and H.J. conceived the ideas of the systematic review and the methodological approach; K.G. and J.S. performed the screening and data extraction of this literature review. Final selection was performed by J.S., C.H.S., K.G., and H.J. The writing was led by J.S., C.H.S., and H.J.