The comparison of caries risk factors in children with very low birth weight and normal birth weight

Health problems of premature infants can affect both general and oral health. The enamel defects, poor dietary and oral hygiene habits may predispose these children to dental caries. Aim Design The study cohort consists from 189 of one year old infants. Anamnestic data were obtained from hospital records, feeding practice, bed-time sugar drinks and oral hygiene onset from questionnaires. Saliva samples of children and their mothers were collected for the detection of cariogenic pathogens.


Introduction
Low birth weight (LBW) is de ned as a birth weight of new born of 2,500 g or less. Very low birth weight (VLBW) is de ned as a birth weight of new born of 1,500 g or less [1]. VLBW is frequently associated with pre-term birth when gestation age less than 37 weeks. Premature and both LBW and VLBW infants have a shorter prenatal period and are at risk to various perinatal and neonatal complications and developmental problems that can affect their general growth and progress during infancy and throughout childhood [2].
LBW is a public health issue because it is closely related to infant mortality and to an increased risk of early and late morbidity, like cerebral palsy, seizure disorders, severe mental retardation, and respiratory tract infections. Premature birth may also affect the developmental and cognitive pro les, IQ scores and perceptual abilities having impact on the child´s behaviour and cooperation during oral health care, dental examination and treatment [2].
Pre-term delivery and LBW can cause severe dental problems like hypomineralization and hypoplasia in both primary and permanent dentition [3,4]. Last trimester of pregnancy is critical for the optimal incorporation of minerals into enamel matrix of primary teeth. Therefore enamel of pre-term children is more porous and less resistant to acidic oral environment. Enamel hypoplasia is more frequently seen on primary incisors like semi-lunar cut of incisal edge [3]. Rough surface of hypo plastic enamel promote plaque formation, and Streptococcus mutans adhesion what may lead to quick progression of caries attack on primary teeth.
The enamel defects with other risk factors as feeding problems, frequent intake of sugar containing medication, oral hygiene habits, early transmission of cariogenic bacteria and behavioural disorders predispose these children to dental caries. Davenport et al. [5] have reported a greater sugar intake, and consequently greater number of dental caries in premature and VLBW children when to compare with full term born children. The recent dental literature has focused on the microbial colonization of healthy fullterm infants and their mothers or caregivers. The literature shows, that the microbial colonization and transmission of cariogenic microbes to the oral cavity of infants and the role of Streptococcus mutans are considered as the risk factor of early childhood caries [6,7]. However, the oral microbial acquisition in pre-term, VLBW infants has not been adequately investigated. We hypothesized that premature, VLBW infants will have different level of cariogenic bacteria in their oral cavity due to differences in the immune system.
The aim of this study was to identify caries risk factors of one year old pre-term, VLBW children and to assess the impact of very low birth weight and prematurity on caries risk in early childhood. Study group including criteria were corrected age one year, birth weight below 1500 grams and gestation age (common term used during pregnancy to describe how far along the pregnancy is; normal pregnancy can range from 38 to 42 weeks) shorter, then 37 weeks. The study group children were included to the project gradually over the period of 12 month as they came for the paediatric examination at the time of corrected age one year.

Materials And Methods
The total number of pre-term, VLBW infants born during the evaluating period of 12 month was 157. Less children from both groups (VLBW, NBW) were nally included to the study for various reasons (child's death, serious health complications, remote residence, parental disapproval of participation in the study).
As pre-term born infants are not fully mature at the time of birth, their chronological age does not correspond to their true biological age. Hence, a comparison with full term infants can only be made if the age of pre-term infants is corrected. The corrected age was calculated by subtracting the number of weeks born before 40 weeks of gestation from the chronological age.
Control group including criteria were chronological age (a measure of an individual's age based on the calendar date on which he or she was born) one year, birth weight above 2500 grams, gestation age longer, then 37 weeks and no systemic disease. These children were selected over the period of 12 month as they came for their rst dental examination at one year of age.
Altogether 189 of one year old infants (102 VLBW and 87 NBW) were included in the study. In two cases there were twins in the study group. Information on their gestational age, mode of delivery, Apgar score (Apgar is a quick test performed on a baby at 1 and 5 minutes after birth. The 1-minute score determines how well the baby tolerated the birthing process. The 5-minute score tells the health care provider how well the baby is doing outside the mother's womb), general health status (congenital heart defects, renal and respiratory disorders, neurological disorders, vision impairments), birth weight and antibiotic use were obtained from hospital records. Feeding practice, bed-time sugar drinks and oral hygiene onset were obtained from questionnaires. All infants participated in the study were residents of communities with low natural water uoride content (< 0.3 ppm). Children were examined clinically at the age of one year (corrected in VLBW, chronological in NBW) by two trained calibrated examiners using the dental mirror and headlight. Examiners were trained to provide treatment using the protocol approved for this study.
Inter-and intra-examiner reliability was evaluated by using kappa coe cients before the start of the study. The standard methodology recommended by WHO has been used (https://apps.who.int/iris/handle/10665/66783). The dental examiners were blinded to the group of infants. The number of erupted teeth (if any part of its crown had penetrated the oral mucosa), developmental defects of enamel (DDE) on the labial surface of the teeth (opacities and hypoplasia) and hard palate deformities were evaluated.

Samples for microbial analysis
Samples for microbial evaluation were taken from both children (91 VLBW and 86 NBW) and mothers (89 VLBW and 86 NBW). In two cases of the VLBW study group there were twins. Some samples were collected in insu cient quantities for microbial and serotype analysis, therefore there were less samples then primary included individuals. Unstimulated saliva samples from the dorsum of the tongue of children and their mothers were collected with a sterile cotton swab. Each swab was transferred into a sterile tube for the detection and identi cation of the main known cariogenic pathogens Streptococcus mutans, Streptococcus sanguinis, Streptococcus salivariu and Streptococcus viridans. Samples for the detection of cariogenic bacteria were inoculated on the selective diagnostic media Mitis Salivarius Agar (OXOID, UK) and incubated 48 hours in 37 degree of Celsius at atmosphere with 10% CO 2 . The main characteristics of the cariogenic bacteria were identi ed (STREPTOtest 24, ERBA-LACHEMA). Serotypes of the isolated Streptococcus mutans were detected using the latex agglutination test Prolex Streptococcal Grouping Latex Kit (BioVendor-Laboratorní medicína a.s, Czech Republic) according the manufacturer instructions. 74 isolates from children and 90 isolates from mothers (both VLBW and control group) were used for the serotyping.

Statistical analyses
Data were described by absolute and relative frequencies of categorical variables and median of quantitative variables. For comparison of categorical variables in groups chi-square test was used, in case of quantitative variables the Mann Whitney U test was adopted, kappa statistics was used for correlation. Analysis was performed in the Stat Graphics Software (Stat Point Technologies, Inc. Warrenton, Virginia, USA), MedCalc v. 9.5.2.0 (MedCalc Software, Belgium) and Microsoft Excel 2016 (Microsoft, USA). Statistical signi cance was set at p < 0.05 two sided.

Results
Anamnestic data of the new-borns (study group VLBW and control group NBW) The total number of 189 infants was included in the study, 102 VLBW and 87 NBW. VLBW new-borns had statistically signi cant shorter gestation age, 29.6 weeks compared to NBW new-borns, 39.8 weeks (p < 0.0001) and signi cantly higher incidence of operation delivery (Caesarean section), 88.2% in comparison to control group, 11.8% (p < 0.0001). The mean birthweight of VLBW children was 1224 grams and they expressed the lower value of Apgar score. The mean birthweight of NBW children was 3315 grams. The differences were statistically signi cant (p < 0.0001) Anamnestic data and oral conditions of one year old children One year old pre-term children signi cantly more suffered from systemic diseases compared to control group (p < 0.0001). The most frequent systemic diseases were congenital heart defects (9.8%), renal and respiratory disorders (7.9%), neurological disorders (2%) and visual impairments (1%). Systemic diseases in NBW children were diagnosed only in 3 cases (atopic eczema, congenital heart defect). Statistical signi cance has been found also in regular medication (p = 0.007). Vitamin D has not been considered as regular medication. Infant characteristics comparison between VLBW and NBW group is presented in Table 1.
VLBW children got more frequently sweetened drinks during the day and night comparing to control NBW children. The difference was statistically signi cant (p = 0.007). Regular oral hygiene practice was more frequent in full term control group, but with no statistical signi cance (p = 0.002). Signi cantly more NBW full term children were breast fed frequently at night when compared to VLBW pre-term children (p < 0.001). This can be explained by the fact that VLBW children were hospitalized for several months at the intensive care units, had less frequent contact with mother, antibiotic therapy and parenteral nourishment.
Signi cant difference has been found between VLBW and NBW children also in number of erupted teeth, when VLBW children had less erupted teeth compared to control group of children (p = 0.036).
Developmental defects of enamel were recorded in both groups, but more frequently in premature children In case of enamel hypoplasia (20.6%) the difference was statistically signi cant (p = 0.033), no signi cance (p = 0.0736) was found in case of hypomineralization (10.8% ). Both hypoplasia and hypomineralization had association with prematurity and children with such defects had signi cantly lower birth weight when compared to children without developmental defects. Oral characteristics and comparison between VLBW and NBW group is summarised in Table 2.

Microbial ndings
Representation of streptococcal species among infants and mothers are presented in Table 3. Proportional representation of S. mutans after the cultivation (MSA + bacitracine) did not reveal statistical difference in both groups of children (p = 0.484). No statistical difference has been found in proportional representation of S. mutans and other streptococci between mothers of full-term infants and mothers of pre-term infants (p = 0.385). Lower proportion of S. mutans was compensated by the higher proportion of S. sanguis and S. viridans.
Almost 40% of isolates S. mutans have shown one from the group anti-gens c, e, f or k. 60% of cases were undetectable. The most frequently detected was serotype c (68.89% in mothers and 67.57% in children), Table 5. The correlation of serotypes of S. mutans isolates has been evaluated and was found 92.52% in serotype c, 87.56% (e), 66.67% (f) and 83.33% (k), Table 6. The correlation of serotypes of S. mutans strains isolated from mothers and their children was κ = 0.832).

Discussion
One year old pre-term children in this study suffered signi cantly more from systemic diseases compared to control group, what may effect orofacial development and oral health. Burt and Pai have reported that pre-term birth and low birth weight are closely related to increased risk of early and late infant morbidity including dental caries in early childhood [8]. Frequent consumption of sweetened drinks, night breast feeding on demand and late onset of oral hygiene practice are considered as other important caries risk factors. VLBW children in our study got signi cantly more frequently sweetened drinks during the day and night comparing to control NBW children. Davenport et al. have reported a greater sugar intake, and consequently greater number of dental caries in pre-term, VLBW children comparing to full term children [5]. Signi cantly more NBW full term children in this study were breast fed frequently at night when compared to VLBW pre-term children. This can be explained by the fact that VLBW children were hospitalized for several months at the intensive care units, had less frequent contact with mother, antibiotic therapy and parenteral nourishment. Regular oral hygiene practice was more frequent in full term control group, but with no statistical signi cance. However predisposition to dental caries in premature children is somewhat controversial, because of the scarcity of studies and necessity to evaluate other primary and secondary risk factors as well as socio-economic consequences [8,9]. One study by Schüler et al. found an association of dental caries with low socio-economic status [10].
VLBW is closely associated with delayed tooth eruption [11]. Viscardi et al. [12] have found that primary teeth eruption occurred signi cantly later in infants with birth weight < 1000 grams and gestation age ≤ 30 weeks. Seow et al. [13] have reported that VLBW infants have less erupted teeth compared to NBW peers at their chronological age 6-17 month. The results of recent study also con rmed the signi cant lower number of erupted primary incisors in VLBW children compared to NBW children.
Developmental enamel defects (DDE) are well studied complications of prematurity and may be present as enamel hypoplasia or enamel hypomoneralization [14]. The possible pathogenesis of DDE in pre-term children may be explained by the presence of many systemic disturbances in prenatal and perinatal period (metabolic disorders, respiratory infections, nutritional disorders as vitamin D and calcium de ciency, birth asphyxia and respiratory distress [3,15]. In our study, enamel hypoplasia has been found in 20.6% and hypomineralization in 10.8% of VLBW children. Enamel hypoplasia with porous and rough surface, combined with feeding problems, frequent consumption of sugar containing diet, behavioural disorders and other risk factors, predispose the pre-term, VLBW children to dental caries in early age [16]. An association of LBW and enamel defects was described by Cortines et al In 2019 [17]. The composition of oral bacteria population in healthy full-term children and their mothers were published. The authors evaluate the microbial colonisation of oral cavity in children, transmission of cariogenic bacteria and the role of S. mutans in aetiology of early childhood caries [18,19]. These factors are insu ciently concerned in pre-term, VLBW children. Makhoul et al. [20] investigated 65 pre-term children and found signi cant rise in bacterial colonization from the rst up to 10th day of life regardless of gestation age and antibiotic therapy. The spectrum of microbial ora is chancing from the 10th day due to antibiotic therapy with the dominance of coagulase negative staphylococci. The intensity of microbial harbouring of oral cavity in new-borns is affected by mucosal immunity system, gestational age, mode of delivery, intensive care unit hospitalization, feeding practice and antibiotic therapy in neonatal period. Based on our ndings, VLBW children had signi cantly more antibiotic medication comparing to control group. Microbial acquisition of oral cavity might be delayed due to surgical mode of delivery (SC) and parenteral nourishment [21,22]. It is in concordance with our ndings. NBW children from the control group section delivery had signi cantly lower prevalence of S. mutans when compared to vaginal delivery.
The early infection of mutans streptococci is associated with caries risk in young children. It was assumed that infants acquire mutans streptococci from their mothers after the eruption of primary teeth.
More recent studies have reported that these bacteria can colonize the mouth of pre-dentate infants. The Swedish study by Nelun Barfod [23] has revealed the 60% detection rate of S. mutans in children aged from 6 to 10 months. Our ndings closely correlate with Australian study using the quantitative RT PCR technique [24]. The differences in detection of oral streptococci may be explained by different laboratory methods used in individual studies.
The studies [24,25] evaluating the oral bacterial colonization in VLBW and NBW children have focused mainly on S. mutans. Wan et al. [6] detected S. mutans in 60% of 6 month old full-term infants and in 50% of the same aged pre-term infants. The same author has found the increasing presence S. mutans with age. S. mutans was detected in 37% of one year old children with higher prevalence in full-term children compared to pre-term children. In this study, S. mutans was detected in 13.19% of VLBW children and in 48.84% of NBW children. Pre-term children had signi cantly lower prevalence of S. mutans and marginally signi cant low grow density. Considering the age of examined children (one year), our detection rate is higher compared to Plonka et al. [26] and lower compared to Merglova et al. [27]. She found the presence of S. mutans in 4.2% of VLBW children and in 100% of NBW children. These observations, coupled with the nding of Fujiwara et al. [28] and Roeters et al [29] clearly illustrate that the presence of S. mutans at one year of age is the effective caries predictor in early childhood. Our study did not reveal differences in S. mutans in VLBW and NDW infants therefore it cannot be concluded that possible higher risk of caries later is attributed to S. mutans. Similarly, Japanese study by Tanaka [30] did not nd any difference in caries prevalence in LBW children. The frequent consumption of sweetened drinks during the day and mainly at night and high salivary levels of cariogenic bacteria in mothers signi cantly contribute to S. mutans colonization of their children [31]. Mothers of both groups of children in our study did not signi cantly differ neither in the prevalence nor in the grow density of S. mutans. The low prevalence and late S. mutans colonization in pre-term, VLBW children can be explained by less frequent contact between mother and infant due to long term hospitalization of these children at intensive care unit, antibiotic therapy, parenteral nourishment and endo-tracheal intubation.

Conclusion
The study con rmed anamnestic and medical differences between one year old pre-term, VLBW infants and full-term NBW infants. The proportional representation and count of S. mutans did not reveal statistical difference neither in VLBW and NBW children, nor in their mothers. Dental health instructions, setting of appropriate dietary and oral hygiene habits, as well as the adoption of e cient preventive measures are of paramount importance in VLBW babies to prevent dental problem in childhood.
Why this paper is important to paediatric dentists Premature VLBW infants are at risk to various perinatal and neonatal complications and developmental problems that can affect both general and oral health. The enamel defects with other risk factors as feeding problems, frequent intake of sugar containing medication, oral hygiene habits, early transmission of cariogenic bacteria and behavioural disorders predispose these children to dental caries.
Dental health instructions, setting of appropriate dietary and oral hygiene habits, as well as the adoption of e cient preventive measures are of paramount importance in VLBW babies to prevent dental problem in childhood.  p=0.3308* VLBW very low birth weight, NBW normal birth weight *Statistical significance determined using ML chi-square test, significant values marked in bold