Streptococcus Mutans Infections in Infants and Related Maternal/Child Factors

Abstract

Objective

The aim of the study was to investigate the maternal-child S. mutans infection, mutual interaction and potential risk factors altering this interaction.

Material and methods

A total of 160 mothers who were paired with their children were recruited. Saliva from mothers and children was collected and S. mutans categorized levels were analyzed. A questionnaire was used to collect maternal and child conventional caries risk factors. A proportional odds regression model was used to investigate whether the level of infection of mothers would similarly affect the level of infection of their children, thus adjusting for other risk factors.

Results

The average age of mothers was 28.1 years and children had a median age of 5.1 months. Paired mother-child S. mutans cross-classification into non-infected and infected subjects showed no agreement in mutual contamination. Maternal infection of the child was highest when the mother was infected with 10^>3 S. mutans/ml of saliva. The results obtained by the regression model showed that none of the potential risk factors influenced the level of infection in the child.

Conclusion

Mothers with a high S. mutans level increase the risk of S. mutans acquisition for her five- month- old child but no evidence was found that other maternal/child factors may influence this.

Introduction

Dental caries is one of the most prevalent dental diseases which requires prevention as a first line approach, (1, 2) especially the form of caries known as early childhood caries (ECC) (3). In children younger than 3 years of age, any sign of smooth-surface caries is indicative of severe early childhood caries (S-ECC) (4).

The consequences of this early form of carious lesions include a higher risk of new carious lesions, an increased cost of dental treatment and time, emergency visits, an increased number of days with restricted activity, later caries in the permanent dentition and malocclusion (2, 5).

S. mutans is the main causative bacterium of human dental caries and plays an important role in etiology of ECC. Moreover, it is crucial in its initial phase (6, 7). The initiation of dental caries is preceded by the colonization of dentition by S. mutans, usually in early childhood (8). The early establishment of these bacteria in the mouth of human infants is mainly dependent on the maternal transmission of saliva (9). The higher the level of S. mutans in the maternal saliva, the more likely is the colonization of primary dentition by cariogenic bacteria (10). The earlier the colonization, the higher is the incidence of carious lesions (11).

This significant correlation between a high caries experience and high S. mutan’s counts has been demonstrated in studies (12, 13). Moreover, the delay of colonization by S. mutans is associated with a lower caries experiences in the future (14). The acquisition of S. mutants may occur very early, even before the teeth erupt (15). Key factors involved in the oral colonization of the cariogenic group of bacteria, S. mutants in young children, need to be studied further (16). Even though several previous studies have tried to document S. mutans mother-child transmission and the related factors, nevertheless, some essential aspects remain elusive (6). Specifically, previous studies (17-21) in this area of maternal S. mutans transmission have shown significant associations using different designs, but this needs further study. Several previous studies looked at qualitative transmission (strains), while our study reports detailed categorical quantitative correlation. The key threshold of S. mutans needed to initiate the caries process is 10⁵ /ml of saliva, however, this needs to be studied further to see how it impacts a mother-child transmission, and more evidence is needed on vertical contamination (19). Another additional issue that needs clarifying is the specific factors affecting maternal-infant early S. mutans transmission. Previous studies (18-21) have explored some of the maternal and older child factors looking for the potential factors mediating S. mutans interaction. To our knowledge, no clinical study has looked into the mother-infant qualitative and quantitative S. mutans detailed interaction or explored specific potential factors that may have an impact on this transmission. Therefore, the aim of this study was to develop knowledge on the maternal-infant S. mutans transmission and identify new caries risk factors that may contribute to this transmission. This initiative is inspired by the medical strategies in ECC prevention such as controlling S. mutans colonization rather than mechanical approaches as suggested by Kohler (20)..

We hypothesized that the specific factors determined for the study would have an impact on maternal-infant early S. mutans transmission.

Methods

Study design and participants

The study design was cross-sectional where maternal/child measurements for the number of S. mutans were performed once. Data collection lasted for a period of 7 months (04-10/2016). A recruitment of 160 voluntary participants was randomly generated from a principal sample of 300 mothers and their young children used for another study that had already been published based on our inclusive criteria (22). The study was carried out in the Main Family Medical Centre (MFMC) of Pristina, the capital of Kosovo. This specific site was considered for data collection since the patients visiting MFMC come from different backgrounds, thus representing a diverse sample for the study.

The inclusion criteria for the study were: if mothers were exposed to the study questionnaire and if the child in the study is under 12 months of age at the moment of saliva collection. The exclusion maternal criteria were receiving chemotherapy or radiotherapy, having a cognitive or physical handicap or having a child being over 12 months of age. Prior to the recruitment of participants, the study was approved by Pristina University Ethics Committee (Ethic Committee-Medical Faculty, University of Pristina, Ref. Nr. 1851) and all mothers who participated it the study gave their consent.

Study Variables:

The principal dependent variable was the child’s levels of S. mutans infection expressed by the categorical and recommended method associated with the commercial test Dentocult SM strip mutans count and as described by Jensen and Bratthall (23).

Bacterial counts were recorded as Colony Forming Units per milliliter (CFU/ml) of saliva. The numbers of bacterial colonies were graded as Class 0, 1, 2 and 3 according to the manufacturers’ scoring-card. Class 0: none detected; Class 1: <10² CFU/ml; Class 2: <10^3-4CFU/ml; Class 3: >10^≥5 CFU/ml.

The principal independent variable was the mother’s levels of S. mutans expressed by the same test and method as their child. Other independent variables of interest were family socio -demographic, economic and behavioral data that were obtained through administrated questionnaires provided for mothers of the children that were being investigated. The information obtained from the mother was set at her level of education, on her last dental visit, knowledge on when to start tooth brushing of the child, on her frequency of daily brushing, on her own oral health perception and on her past caries experiences. The information obtained on the child was the presence or not of his/her teeth and the number of times he/she has been ill. The examiner was asked to observe the mother and to check the upper anterior teeth to see if there was any dental plaque or clinical changes relating to past or present caries experiences.

Collection of the variables: Saliva collection was carried out using the method by Dentocult SM stripmutans test (23) using a plastic strip. All samples of saliva were collected by a dentist or a PhD candidate. Maternal questionnaires were also administrated by the same examiner.

Data analysis: For each categorical and continuous variable of interest, the counts and percentages were reported, as well as the median and inter-quartile range (IQR=75^th percentile–25^th percentile). To investigate the association between maternal S. mutans infection and their child a S. mutans acquisition the McNemar’s test was used (24), and the kappa coefficient of agreement between the categories was calculated. Also the proportions of infected children within the cohort of infected mothers were compared. To investigate the marginal effect of the level of S. mutans infection in the mother on the outcome of the level of S. mutans infection in the child a regression model was used to take into account the ordinal nature of the coding for this outcome. The regression model was adjusted to take into account other risk factors of interest (Table 1), the proportional odds or cumulative logits. The results were reported as odds ratios (OR) with a 95% confidence interval (CI).

Table 1. Distribution for categorical mother/child demographic variables.

	Categories	n (%)
Mother mean age	Years (26-31)	28.14
Child median age	Months (4-12)	5.01
Mother level of education	Graduated from high school or less?	73 (45.6%)
	Graduated from College or University?	87 (54.4%)
Family income (monthly)	Less than 500 €	77 (48.1%)
	600-1,000 €	66 (41.3%)
	1,000-3,000 €	17 (10.7%)

All statistical tests of the hypothesis were two-sided and performed at a significance level of 5%. All statistical analyses were done using the SAS software, version 9.3 (SAS Institute Inc., Cary, NC, USA) (25).

Results

The study included 160 mother/child pairs. The median age of the mothers was 28.0 years (IQR 26-31) and the median of the children was 5 months (IQR 4-12, in that last case, due to a skewed distribution, we used median instead mean). The results showed that more than half of the mothers (54.4%) had graduated from university and almost half of them (48%) had a low family income. Almost all mothers were infected by S. mutans (150/160, 93.8%). The rate of infection for the infants was 70.6% (113/160).

A cross-classification was performed for the mother-child dyad for the presence of infection (Table 2). It was found that children of infected mothers would not necessarily be infected and that children of non-infected mothers would also be non-infected, i.e. the off-diagonal counts were not very small (McNemar’s test p<0.0001). This result was validated by a very small kappa coefficient 0.04. However, in the group of infected mothers, the proportion of infected children was 71.3% (107/150), with a 95% CI (64.1%, 78.6%) (Test for proportion, Z=5.23, p< 0.0001). The proportions were not significantly different in the group of non-infected mothers.

Table 2. Mother-child dyads according to infection status.

	Child			P-value
Mother	No	Yes	Totals	<0.0001*
No	4	6	10
Yes	43	107	150
Totals	47	123	160

* McNemar’s test

When the level of infection of the mother/child dyad is considered, a cross-classification shows the same level of infection (class 1-10²) present in 22 of the mother/child dyads. This shows that there is a very low correlation between the level of infection categories on the mother/child dyad (McNemar’s test, p< 0.0001; weighted kappa = 0.05).

We then performed an additional analysis and obtained the results from a proportional odds regression model, which was later used to investigate whether the level of infection of the mother would similarly affect the level of infection of their children, adjusting for other factors. The results showed that none of the risk factors influenced the child’s level of infection.

Discussion

The results of this study have shown that 70.6% of the young children were infected by S. mutans. Similar results were obtained in a study conducted by Damle et al. (2016) (25), 77.3% of young children were found to have S. mutans, up to 30% of the 3-month-old children were infected, whereas children who were 6 months old had an infection rate which increased to 60%. On the contrary, other studies have shown different results for S. mutans infection; in children 6-18 months old the infection rate was found to be 27%, 30% and 53% (26). S. mutans can be found at early ages: Milgrom et al. (27) found the prevalence to be 53% in 6 to 12-month-old children, although Karn et al. found the prevalence in 15-month-olds was 60% (28).

The high level of S. mutans infection was also found to be present in mothers who were included in the study. The result of 93.8% of infected mothers with bacteria shows a very high rate of infection. Li et al. (2000) also reported a high level of infection of 87.5% among mothers (29), whereas in a pilot study (n=10) S. mutans was found in all mothers (23).

One of the study objectives was to report on S. mutans infection in infants taking into consideration the level of maternal infection. The results illustrate a true dose-response between the high maternal S. mutans infection rate, 10³–10⁵ organisms per milliliter of saliva, and the subsequent child S. mutans colonization rate (63.7%), whereas in low maternal S. mutans infection, 0-10² organisms per milliliter of saliva, only 36.2% of children had a S. mutans infection. From the literature review, no studies were found which investigated the mutual S. mutans levels of infection in mother and child at such a young age. A previous comparable study by Berkowitz et al. (1981) in children who were 10-16 months old found that the frequency of infant infection was approximately nine times greater when maternal salivary levels exceeded 10⁵ organisms per ml (30).

When only the qualitative presence or absence of mother-child infection was taken into consideration, it was found that in the group of infected mothers the proportion of infected children was 71.3% (107/150). This showed that not all infants of infected mothers would also be infected and also not all infants of non-infected mothers would also be non-infected. When the intensity quantitative level of mutual S. mutans infection is considered, from reported associated frequencies (Table 2) or categories (Table 3), a significant linear correlation construction was not found between both distributions. Previous studies on older children (31, 32) confirmed a quantitative correlation. In general, S. mutans infections in children either by primary sources (33) or time expression (14, 33) have been reported as very varied. This may explain why no symmetric S. mutans distribution was found. Also, other studies have depicted that S. mutans is readily acquired from nonmaternal sources in certain populations (34).

Table 3 S. mutans child infection associated to mothers one.

Mother S.mutans levels	Child s. mutans levels in CFU/ml of saliva				Total
	0	102	103-4	10≥5
0	4	1	4	1	10 (6.2%)
102	14	22	8	4	48 (30.0%)
103-4	18	20	12	4	54 (33.7%)
10≥5	11	23	5	9	48 (30.0%)
Total	47(29.4%)	66(41.2)	29(18.1%)	18(12.1%)	160

Another objective of the study was to assess S. mutans infection between the mother and her infant in detail and to evaluate the potential risk factors that may impact on this. It is documented that S. mutans from either the mother or caregiver is the main source for early child acquisition (32). The most potential ECC and S. mutans transmission risk factors are partially documented from previous studies: maternal socio-demographic and economic status, oral hygiene practices (35), maternal clinical observations and caries experiences (16, 35) child dietary behaviors (18, 20, 36, 37) teeth eruptive timing, breast-feeding duration (29) child enamel defect and low birth weights (36, 37) and child sex (36). Therefore distinctive or complementized factors of previous studies based on our review were selected considering notably the age of participants. The factors studied were from maternal demographic, behavior, knowledge and clinical observation collections and child illness episodes. The starting point was first exposure to tooth brushing and tooth presence.

The limitation of our study was a relatively small sample size. The high costs of laboratory analysis and tests were unaffordable; hence we could not include a greater number of participants in the study.

Conclusion

None of the factors analyzed influenced the impact of maternal infection on their child’s S. mutans outcome. Regarding the evaluation of potential factors, a proportional regressive model was used to document whether the maternal S. mutans levels of infection would similarly change their child’s degree of infection, thus adjusting for other risk factors that were collected.