Association of early childhood caries and nutritional status: a scoping review

2.1 Search strategy

The study methodology of this scoping review was based on the six-stage methodological framework outlined by Arksey and O’Malley that consisted of “identifying the research question, identifying relevant studies, study selection, charting the data, collating, summarizing, reporting results and consultation (optional)” with consideration of the scoping framework suggested by Levac et al. [28, 29]. The research question for this scoping review is as follows: “What is the association between ECC and nutritional status?”. A search of published literature until December 2022 was performed on PubMed, Scopus, and Web of Science databases using the following search terms (“Early Childhood Caries*” OR “Dental caries*” OR “Caries*”) AND (“Nutritional status*” OR “Nutrition status*” OR “Nutritional assessment*”) AND (“Growth*” OR “Growth and Development*” OR “Development*”) and the search was restricted to articles in English.

2.2 Criteria for study selection

The inclusion criteria for the selection of articles were defined according to participants, concept, and context domains (PCC): Participants (P): Children aged 71 months and below regardless of gender/race; Concept (C): Children assessed for both ECC and nutritional status (anthropometric measurement, biochemical status such as vitamin D, clinical assessment, based on dietary, or environmental/psychosocial data); Context (C): Dental assessment, nutritional assessment, other factors and the association outcome. From the search results obtained, review and systematic review articles were excluded. Articles involving animal studies, sample population of children aged more than 71 months old, and articles not in English language were excluded.

2.3 Data extraction

The initial search yielded 492 articles. Screening of titles and abstracts was conducted by two independent reviewers (LDT and FY). A total of 56 duplicate articles and 226 irrelevant articles were removed using an online platform called Rayyan (https://www.rayyan.ai/). In case of disagreement between reviewers, a discussion was conducted until a consensus was reached. A total of 210 potentially eligible full articles were considered for full text evaluation. However, 190 articles were excluded based on the inclusion and exclusion criteria. The excluded articles consisted of 122 unrelated articles, 59 articles with wrong population of children (aged above 71 months), and nine systematic review or review articles. A total of 20 articles were included in this scoping review. A flowchart of the selection process according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) is shown in Fig. 1. Data from the 20 shortlisted articles were extracted and tabulated in Microsoft Excel 365 according to the author, year, sample population, dental assessment, dental outcome, nutritional assessment, other factors, and association outcomes, as presented in Table 1. The quality assessment of the included articles was appraised using the adapted version of Newcastle-Ottawa Scale (NOS), as presented in Table 2 [30].

Fig. 1.Flowchart according to the PRISMA checklist showing the flow of articles arising from the main search.

3.1 Type of study, sample population and quality assessment

As shown in Table 1, there were five types of studies among the shortlisted articles. Out of the 20 studies, thirteen studies were cross-sectional studies [22, 23, 24, 26, 31, 32, 33, 35, 36, 37, 38, 39], one retrospective study [40], one subcohort cluster randomised controlled trial [27], one case control study [41], two cohort study [25, 44], one longitudinal twin study [43], and one longitudinal study [42]. The subcohort cluster randomised controlled trial study by Muhoozi et al. [27] involved 203 children in the intervention group who received education intervention on oral hygiene practice and dietary advice, while the other 198 children in the control group did not. However, there was no data regarding caries in relation with nutritional status [27]. All the studies had study populations that included healthy children aged 71 months and below, as defined by the American Association of Paediatric Dentistry (AAPD) for children diagnosed with ECC [3]. There was a great variation in respect to sample population used in the studies ranging from as low as 46 children to as high as 27,333 children [23, 40].

In terms of quality assessment, nine studies [22, 25, 26, 27, 32, 34, 38, 41, 44] were considered high quality, and the remaining eleven studies [23, 24, 31, 33, 35, 36, 37, 39, 40, 42, 43] were considered moderate quality due to insufficient descriptions on the sample size calculation and methods for controlling confounding factors such as socioeconomic status, diet, and oral hygiene practice. Detailed information regarding quality assessment is tabulated in Table 2.

3.2 Dental assessment and dental outcome

A total of eleven studies used dmft or deft index [22, 24, 25, 31, 33, 36, 38, 40, 41, 42, 44] as their dental assessment for ECC. Two studies used dmfs index [31, 34] while three studies used dft index [32, 35, 37] as their dental assessment index where “d” represents decayed teeth, restored teeth with secondary caries or caries on another surface while “f” represents teeth with restorations but no secondary caries as it gives a good understanding of the tooth decay status [35]. Missing teeth were not included as there was no definitive aetiology of whether the teeth underwent early extraction or did not exist [35]. Next, ICDAS scoring was used by two studies [39, 43]. Silva et al. [43] used the ICDAS scoring to categorise their samples into “any” or “advanced” caries, while Cuong et al. [39] used the scoring method to diagnose the presence of caries. Decayed teeth (dt) assessment [23, 24, 26, 32] was used in four studies, while another study used close-up intraoral photographs of upper front teeth of the children to analyse the presence of carious lesions [27]. Only obvious cavitated lesions into the dentine were considered carious [27].

Dental outcomes reported in the studies consisted of indexes such as dmft index and caries prevalence. Caries prevalence was reported in twelve studies with 82.3% being the highest [31]. Out of those studies, two studies with high caries prevalence [31, 39] and seven studies with low caries prevalence [22, 25, 32, 34, 35. 37, 40] showed a positive association between ECC and nutritional status, while three studies [26, 27, 43] showed no association. All the studies reported dental outcomes based on the indices used, as shown in Table 1, except for three studies [23, 41, 42].

3.3 Nutritional assessment

Nutritional assessment was carried out in all the selected studies. Anthropometric measurements were used in all studies except for two studies that used biochemical assessments such as vitamin D, haemoglobin, mean corpuscular volume, serum ferritin, and serum albumin levels [23, 41]. The anthropometric measurements obtained for weight, height, and calculated BMI were converted using the WHO Anthroplus software (Version 1. 0. 4, WHO, Geneva, Switzerland) to categorise the children into categories of stunted, underweight, normal, overweight, and obese respectively. Undernutrition children with HAZ and WAZ less than −2SD were classified as stunted and underweight accordingly while WHZ of less than −2SD indicated wasting. Severely undernutrition children were scored less than 3SD below the median [45]. As for BAZ, overweight and obese were classified in children aged 60 months and below with BAZ above +2SD and for children aged above 60 months with BAZ ≥ +1SD [45, 46]. SD stands for standard deviation. To assess blood biomarkers, venipunctures were conducted for collection of blood samples [41].

3.4 Other factors

The selected studies showed variations in the additional data collected, besides dental and nutritional outcomes. These parameters have the potential to affect the association between ECC and nutritional status. The data regarding sociodemographic factors and socioeconomic factors were collected using questionnaire in thirteen studies [22, 24, 25, 26, 31, 32, 34, 36, 37, 38, 40, 41, 44] and twelve studies [22, 24, 25, 31, 32, 34, 36, 37, 38, 39, 41, 43], respectively. Dietary intake assessment, sugar intake, as well as cariogenic food frequency and exposure were conducted in eight studies through diet recalls [22, 25, 26, 33, 37, 39, 41, 43]. Three studies also further attained information regarding dietary habits by observing feeding practices [27, 36, 38]. A small proportion of the studies obtained the data regarding community water fluoridation through sampling of water at the study area [27] or obtained the available data from the local government [40, 43]. Oral health status and practice were evaluated in five studies through questionnaires based on the frequency of brushing, assistance during brushing [22, 32, 39, 41, 43], and through examination of visible plaque [35]. Other factors that were also present in the selected studies were prenatal history, drug history of mother, developmental milestones [36], second hand smoke exposure [26], use of supplements [33], sun exposure and skin pigmentation [41].

3.5 Outcome/findings

The outcome of the selected studies showed that ten studies reported an association between ECC and poor nutritional status children who were stunted or underweight [22, 23, 24, 31, 34, 36, 38, 39, 41, 44]. A general association between ECC with both undernutrition and overweight or obese children was reported in five studies [25, 32, 35, 37, 42]. One study reported an association between ECC and higher BMI children who were overweight or obese [40]. No association between ECC and nutritional status was reported in four studies [26, 27, 33, 43].

4.1 Dental assessment

There were various dental assessments used in the studies included in this scoping review, namely dmft, deft, dft, dmfs, ICDAS scoring and photographs of anterior teeth. Based on the WHO diagnostic criteria for the diagnosis of dental caries, dmft is defined as the sum of the number of decayed, missing due to caries and filled primary teeth [47]. Majority of studies used the dmft index as a dental assessment and an outcome measurement for dental caries as this index is one of the simplest ones to use [48]. However, since dmft includes both untreated decay and filled tooth/teeth due to caries, population with high rate of untreated decay and population with high rate of filled tooth will have similar high dmft values [38]. The severity and chronic nature of the decayed teeth may be masked when the dmft index is used to determine the present and past caries experience [26]. Thus, Lee et al. [26] suggested the number of decayed teeth (dt) as a better or stronger determinant than dmft [49]. Based on the WHO oral health survey, dt index is a criteria to score dental caries, besides dmft, but WHO does not state the superiority of one over the other [5]. The diagnostic criteria of dt includes a lesion in pit, fissure, or on a smooth tooth surface a detectably softened floor or wall, undermined enamel, filled tooth with decay, or an obvious cavity [26]. The number of untreated caries in the oral cavity is indicated by the number of decayed teeth [50].

Besides that, another form of dental assessment used was close-up intraoral photographs of upper front teeth where obvious cavitated lesions into the dentine were considered as presence of caries [27]. The study by Ardenghi et al. [51] showed that the development of caries in the maxillary anterior teeth of primary teeth was associated with the development of caries in posterior teeth. This can also be used as a good predictor of caries in permanent dentition. However, compared to dmft, photographs of upper front teeth may not represent the condition of the entire dentition and may lead to underdiagnosis [27].

4.2 Nutritional assessment

According to Norberg et al. [24], to better understand the relationship between dental caries and body weight, dental caries should be studied in relation to the whole span of weight or BMI. There are many factors that affect BMI, such as carbohydrate consumption, socioeconomic status, dietary pattern, oral hygiene practice, and alterations due to genetic aspects [52, 53]. Body mass index for age has been proven to be a sensitive indicator of body fatness in children [54]. Majority of the articles in this scoping review used the classification by WHO Child Growth Standards for children aged 60 months and below and WHO Growth Reference for children 60 months above that uses the HAZ, WAZ, and BAZ to categorise the children’s nutrition status [45, 46]. The WHO Growth Standards describes the growth of a healthy child who is fed according to the WHO feeding recommendations. Any deviation of a child’s growth from the standard indicates abnormal growth [55]. As for the WHO Growth Reference, it describes “the growth of a sample individuals and provides basis for comparing populations without making inferences about the meaning of observed difference”. This is used as a reference but not a standard [55].

Other than that, HAZ, WAZ, and BAZ are z-scores that describe “the number of standard deviations (SD) of the actual height, weight or BMI of a child respectively, from the median value of his/her age as determined from the reference population” [56]. The z-score is recognised as the best system for analysis and display of data compared to other methods because it is quantified based on the distribution of reference population. Thus, the score reflects the reference distribution and standardised quantities allowing it to be comparable across age, sex, and anthropometric measures [57, 58]. However, there may be differences between study population and the reference population used in the calculation of z-scores [38].

Laboratory methods such as measurement of vitamin D, serum ferritin, and serum albumin levels can be used to determine nutritional status [23, 41]. Vitamin D plays an important role in the formation of enamel and dentin, as well as maintenance of good oral health and craniofacial development [41]. Vitamin D is also an important component in bone mineral metabolism and skeletal growth and development [59]. Vitamin D supplementation has been associated with reduced caries risk and prevention of caries [60]. According to the study by Schroth et al. [41], caries-free children were twice as likely to have an optimal vitamin D concentration, while those with severe ECC have nearly three times the odds of having deficient levels. In addition to low vitamin D levels, unsatisfactory levels of albumin, haemoglobin, serum ferritin, calcium and parathyroid hormones are indicators of nutritional deficiency or malnutrition [23, 41]. In the relationship between ECC and nutritional status, the chronic dental infection may cause effects to the systemic body response which may include low vitamin D levels that affect growth [61]. Low levels of ferritin and haemoglobin cause iron-deficiency anaemia that can affect children’s brain development, body function, and growth [62]. A study by Clark and team showed that severe ECC may be a risk marker for iron-deficiency anaemia [23]. Laboratory methods, however, poses a disadvantage during phlebotomy procedure on children. The procedure is an invasive investigation and time-consuming, unlike the anthropometric measurements, which are low cost and simple to conduct [15].

4.3 Other confounding factors

Factors that may relate to the relationship between nutritional status and ECC are the socioeconomic status and sociodemographic status such as gender, household income, and parental education level [53, 63]. Lee et al. [26] reported that males in general had a higher number of decayed teeth compared to females due to psychological factors such as noncompliance, faster eruption, and retentive primary teeth. On the other hand, two studies showed no differences between males and females [24, 52]. Various studies showed that the presence of caries was higher in children with a low socioeconomic status [53, 64]. The study by Renggli et al. [43] showed that children, predominantly those in the rural provinces, had below par baseline anthropometric measurements, dietary findings, and caries experience compared to the urban provinces. This may impact the formation of teeth by causing dental defects such as brittle and poorly mineralised dental enamel, which is a risk factor for caries. On the other hand, due to the high and excessive sugar consumption among the urban children, their caries experience worsened over time and may confound any potential causal relationships between caries and nutritional status [44]. In addition to that, underweight children in low- and middle-income countries and overweight children in high income countries were reported to have high levels of dental caries [63, 65].

Poor oral hygiene, together with other risk factors for caries, has been associated with caries formation [66]. One of the studies used presence of visible plaque to help explain the lack in oral hygiene practice and parents support in toothbrushing [22]. The start of toothbrushing behaviour before the age of two years old has been related to being caries-free at four years old [67]. Based on AAPD, brushing with fluoridated toothpaste twice daily is recommended for all children for prevention of caries [68].

The study by Lee et al. [26], highlighted the positive association between second-hand smoke exposure on the formation of dental caries. Children had a 1.72 times higher risk of having caries in their primary teeth if they were exposed to second hand smoke during infancy [69]. The second-hand smoke increases the Streptococcus mutans levels and decreases the mineralisation of tooth and rate of salivary flow, thus leading to an increased risk of caries formation [70].

4.4 Effect of early childhood caries on nutritional status

ECC can have a direct impact on children’s nutritional status, growth, and development as the disease causes a reduced intake of food and nutrients due to the pain and discomfort when eating [26, 71]. The insufficient food consumption to cater for the metabolic needs of children will affect their growth [22, 36]. ECC can have detrimental effects to children’s immediate and long-term quality of life [72, 73]. The effects of caries on children have been linked to failure to thrive, impaired speech development, and odontogenic infection with premature loss of primary teeth that may affect the development of permanent dentition [27, 72]. Children with severe dental caries have been shown to have double the risk of developing chronic malnutrition [44]. Weight gain and catch up growth were seen in underweight children after dental treatment of caries, revealing that caries can hinder growth and development [74]. Pain and discomfort from toothache may cause disturbed sleep, which may affect glucosteroid production and growth [75]. Metabolic pathways involving cytokines may also be affected by the presence of chronic inflammation due to pulpitis and dental abscess [36].

4.5 Effect of nutritional status on early childhood caries

Nutritional status either underweight, normal, or overweight is related with dietary intake. This depends on the total amount of calories consumed especially from calories that derive from sugar consumption [25]. The amount of sugar consumption can lead to overweight and obesity, while the frequency of sugar consumption is a risk factor for caries formation [76]. Many previous studies have supported this positive association where high frequency of both liquid and solid sugar intakes leads to a high caries risk [22, 26]. This association is supported by Stephan’s curve which shows that a fall in pH below 5.5 (critical level) will cause demineralisation of enamel [77]. Consumption of fermentable carbohydrates or sugar reduces the salivary pH beyond the critical pH value. Gradual return of pH takes 30–40 minutes before remineralisation can take place [78, 79]. Prolonged low pH due to repeated intakes of carbohydrates does not allow remineralisation to happen [77]. Thus, a high frequency of sugar consumption increases the episodes of drop in pH values, which then increases the risk for caries formation [26]. Frequent night feeds lead to formation of dental caries due to the physiological reduced salivary flow whilst asleep [27]. Bottle feeding among preschool children has also been proven to be significantly associated with ECC [80].

In terms of the aetiology of obesity, besides genetic factors and frequent consumption of high calorie and sugary food, children’s food preference also plays a very important role [24, 44, 81]. Children are generally picky eaters when they have a strong food preference, inadequate variety of food groups intake, reluctance to try new food and eating lesser amount of food [82]. In addition, children may have undernutrition issues and dental caries as the nutritious food and drinks that are needed for growth have been replaced by consumption of sugary foods and drinks [44]. According to Dubois et al. [83], picky eaters were twice as likely to be underweight at the age 4.5 years compared to the non-picky eaters.

Obesity may lead to increased susceptibility to caries as obesity is related to the salivary flow and immune response [40]. A correlation was discovered by Gerdin et al. [64], who found that obese preschool children may have a risk of developing dental caries at a later age. However, the inability to eat adequately due to caries or the presence of undernutrition due to wasting may also affect salivary composition and flow, thus leading to more caries [38].

4.6 Association between early childhood caries and nutritional status

There is vast literature on the relationship between caries and nutritional status [44]. The differences in the literature may be due to the complex relationship between caries and nutritional status, which may be affected by many factors such as socioeconomic status, child age, oral hygiene practice, and county or location (urban vs. rural) [44, 53, 84]. According to the systematic review by Chen et al. [65], there is a high occurrence of caries among overweight and obese children in high-income countries, but not for children from low- or middle-income countries. Besides that, the association between ECC and overweight or obese children is also supported by Ribeiro et al. [25]. However, there are also studies that reported no association between overweight/obesity and dental caries [24, 85]. The children with low BMI were found to have more caries than children with normal weight, and this was proven to be statistically significant [24]. This statement is also supported by various other researchers [25, 38, 86]. Other than that, previous studies also showed no relationship between ECC and nutritional status [26, 43, 52]. This may be due to the chronic nature of both ECC and nutritional deficiency such as obesity or stunting, that may take several years to develop. Thus, the effects of these diseases may be more obvious in older children with permanent dentition [26, 87]. This may eventually cause inconsistency in the association [52, 88].

In terms of the effects of low and high caries prevalence population on the association of ECC and nutritional status, there were conflicting results between the shortlisted studies. High caries prevalence population can affect the association between ECC and nutritional status. In two studies with high caries prevalence population, the association was seen between ECC and stunted children [31, 39]. High caries prevalence population showed that the caries occurrences disturbed the children’s ability to eat, thus causing a poor dietary intake that affected the growth and development [31]. On the other hand, the association between ECC and nutritional status was found in seven studies involving low caries prevalence populations. Majority of the studies had a general association between malnutrition and ECC, while one study associated ECC with underweight and another study associated ECC with higher BMI [22, 40]. This variation did not show a clear effect of low caries prevalence population on the association of ECC and nutritional status. This may be due to other confounding factors that may have affected the association, such as socioeconomic factors including family income and parent’s education [34]. Therefore, there were differences in the association between ECC and nutritional status in high and low caries populations.

4.7 Strength and limitations

The strength of this review was the inclusion of methodological quality assessment by using the Newcastle-Ottawa quality scale to provide information regarding the strength of the included studies [89]. The high-quality studies included data regarding justification of sample size calculation, assessment outcome, and supporting statistical tests with sufficient statistical power, where moderate-quality studies lacked these criteria [22, 25, 38, 41, 44]. The type of study was also a limitation, as majority of the studies included were cross-sectional studies. A cause and effect relationship cannot be determined by a cross-sectional study [52]. A cohort study may be more suitable when establishing relationships between factors [26]. One interventional study compared groups with and without oral hygiene and dietary advice but did not include information regarding caries in different groups of nutritional status such as undernutrition, normal, and overnutrition [27]. Hence, the cause-and-effect relationship between ECC and nutritional status cannot be determined.

Besides that, the confounding factors in the shortlisted studies, such as sociodemographic profiles and socioeconomic factors were not controlled except in one study by Silva et al. [43], which involved participation of twins. A longitudinal twin study may have advantages as it removes the bias resulting from confounding factors as twins have similar genetic variation, and this allows adjustment of known and unknown potential shared confounders [43]. Confounders have also been found to modify the outcome of the association between ECC and nutritional status [90]. There was also heterogeneity of the dental and nutritional assessments used. The dental assessments consisted of dmft, deft, dft, dmfs, ICDAS scoring, and photographs of anterior teeth, while nutritional assessments consisted of weight, height, HAZ, WAZ, BAZ, and BMI. Thus, there was a lack of homogeneity among the studies included. These differences and lack of consistency in the studies were the limitations of this review and may influence the analysis of the results. This review also presented the results in a descriptive manner, unlike a systematic review that may be more analytical. Therefore, further well-designed studies are needed with similar sample population and consistent methodology to allow comparability to yield more valid conclusion.