Maternal Oral Mutans Streptococci (MS) Status,Not Breastfeeding,Predicts Predentate Infant Oral MS Status

Introduction

The role of Streptococcus mutans and Streptococcus sobrinus (two closely related bacterial species, together referred to as mutans streptococci [MS]) in dental caries was first identified in 1924, ¹ and in 1986, the often-observed link between oral S. mutans bacteria and dental caries was demonstrated. ² Viable MS can be recovered from inanimate objects, such as a metal plate up to 24–48 hours after contamination, and MS is also transmittable directly through saliva. ³ In the 1970s, Berkowitz et al. ⁴ described the similarity of oral MS presence in mother and infant, and fidelity in terms of similarity of molecular fingerprint of S. mutans in the mother and her infant was demonstrated in the mid-1990s. ⁵ Differentiation of S. mutans from S. sobrinus and positive detection for S. mutans were demonstrated using arbitrarily primed polymerase chain reaction fingerprinting in the late 1990s.^6,7 In 1999 Kononen et al. ⁸ showed that once oral anaerobes were established in the oral cavity as early as 2 months of age, they tended to persist. A review of published studies concerning sources of colonization of MS in children concluded that mothers appear to be the primary source, ⁹ and breastfed infants were found at higher risk for oral MS compared with bottle-fed infants.^10,11

As an ancillary investigation within an ongoing clinical trial, we had the opportunity to examine oral MS status in 4–6-week-old predentate infants and their mothers. Therefore, the purpose of this study was to investigate the association of infant feeding method (breastfeeding versus formula feeding) with infant MS status, in the context of mother–infant pairs.

Materials and Methods

Ethical clearance for this study was obtained from the institutional review board at the Medical University of South Carolina. Mothers had granted informed consent in English or Spanish.

Data for this analysis were obtained from an ongoing prospective clinical trial of mother–infant pairs, in which all infants were more than 36 weeks of gestation at birth and the mothers had self-selected the feeding mode for their infant. The breastfed infants were allowed up to 10% formula usage, whereas the formula-fed infants were exclusively formula-fed. When the infant was 4–6 weeks old and before any intervention on the clinical trial, a chairside test for oral MS was performed on the mother and on the infant, using the Dentocult^® SM Strip mutans test kit (Orion Diagnostica, Espoo, Finland) for detection of salivary MS. This kit was introduced in 1989 with reported sensitivity and specificity of 98% and 85%, respectively, and a positive predictive value of 93% and negative predictive value of 94%. ¹² Oral sampling was performed from unstimulated saliva in both mother and infant. After incubation, the presence of colony-forming units (CFU) of MS per milliliter of saliva was scored from 0 to 3 according to the manufacturer. Negative status (score of 0) was indicated by <10⁴ CFU/mL. Positivity for MS was indicated by a score of 1 (≥10⁴ and <10⁵ CFU/mL), 2 (≥10⁵ and ≤10⁶ CFU/mL), or 3 (>10⁶ CFU/mL). Inter- and intra-examiner agreement for three examiners reading the MS tests was 100%. Demographic and perinatal data for the mother and infant were abstracted from the parent study database. REDCap software was used for data entry. Tabulations, chi-squared test, Fisher's exact test (if any cell was less than 10), and logistic regression modeling were performed using STATA IC version 11.0. ¹³

Results

From 111 mother–infant pairs who had MS status tested at the 4–6-week visit, complete data were available for 104 and are the subject of this analysis. Demographic data (Table 1) did not differ significantly if all 111 mother–infant pairs were included (data not shown). Among the 104 infants, 68 were breastfed, and 36 were formula-fed. Mother's race/ethnicity distribution was Hispanic (61%, n=63), white (22%, n=23), and African American (15%, n=16), and two pairs were classified as other (3%). The distribution of the mother's race/ethnicity differed significantly between the two feeding groups (p=0.006), with whites and Hispanics more likely to breastfeed and African Americans equally likely to breastfeed or use formula. However, the two feeding groups did not differ significantly by delivery mode (vaginal or cesarean), by level of care at birth (newborn nursery versus higher level of care [e.g., Level II]), or by the presence of MS in the mothers or in the infants. Within each feeding group, the presence of MS in the infant's saliva did not vary significantly by mother's race/ethnicity (10–15% of infants were positive for MS in each ethnic/racial category; data not shown).

Overall, 31% (n=32) of the mothers and 12% (n=12) of the infants were positive for salivary MS (Table 2). In total, 69% (72 pairs) were congruent between mother and infant, and in 6% (six pairs) both mother and infant were MS positive. Infant oral MS status was not predicted by race of mother, by delivery mode, by feeding method, or by mother's salivary MS status (Table 3). However, when stratified by feeding method, the mother's salivary MS status did significantly predict the infant's salivary MS status in the breastfed group (odds ratio=5.97; 95% confidence interval, 1.06–33.7; p=0.043) but not in the formula-fed group (odds ratio=0.72; 95% confidence interval, 0.07–7.42; p=0.782). In multivariable logistic regression modeling this relationship was further modified by race of mother, but only among breastfed infants, where mother's MS status predicted the infant's MS status with odds ratio=8.53, p=0.027 if race was included (data not shown).

Discussion

Our multi-ethnic study of 104 mother–infant pairs from coastal South Carolina found that 12% (12/104) of full-term infants were MS positive in saliva at 4–6 weeks of age using the Dentocult SM test. We found infants in both our breastfed (10%, 7/68) and formula-fed (14%, 5/36) groups were positive for MS at 4–6 weeks of age. We further found that salivary MS status of the infants was predicted by the mother's salivary MS status, but only if the infant was breastfed and not among formula-fed infants. That is, neither feeding method nor mother's salivary MS status alone was significantly associated with the infant's MS status. A comparable study found 9% (25/283) of Australian infants positive for MS at 5 weeks of age using the chairside diagnostic kit CRT^® S. mutans test (Ivoclar, Melbourne, Australia). ¹⁴ Positive maternal MS status (>10⁵ CFU/mL), but not breastfeeding alone, was significantly associated with MS in the infant. ¹¹

Using a plated culture technique for S. mutans detection, a relatively higher proportion of full-term Australian infants, 34% (44/128), was S. mutans positive at 11–15 weeks ¹⁵ compared with our findings. Again, in a similar cohort of predentate infants, 60% (67/112), a higher proportion was positive for S. mutans at 6 months of age; the authors attributed the high proportions to their sample size and sensitivity of their culture media. ¹⁰ Likewise, the proportions of predentate infants found to be S. mutans positive in the above studies are higher than in a recent study in a subset of 73 infants at 3 months of age from a small town of Northern Sweden, ¹⁶ tested using the Human Oral Microbe Identification Microarray. Although Streptococcus Cluster III (which included S. sobrinus) was detected in >90% of samples, only 6.8% (5/73) of breastfed and 0.0% of formula-fed infants were positive for S. mutans.

In the above studies of predentate infants, there are differences in infant age (34 days to 15 weeks and then 6 months), geographic location (Australia, Sweden, and the United States), and most notably the methods for species detection and the grouping or not of specific species of the mutans group. For example, at one extreme, 1–10³ CFU/mL S. mutans was considered positive using the plated culture method,^10,15 whereas both the Dentocult SM test and the CRT S. mutans test use a cutpoint of >10⁴ CFU/mL for positive MS status. A recent study by Chaffee et al. ¹⁷ used a plating and culture method for MS (S. mutans and S. sobrinus) and an a priori cutpoint of ≥4.5×10³ CFU/mL for higher MS challenge. That cutpoint was based on an earlier study of Japanese mothers' whole saliva specimens for S. mutans and S. sobrinus that indicated that ≥4.5×10³ CFU/mL was predictive of both MS colonization and prevalence of dental caries in children at 2.5 years of age. ¹⁸ With our higher detection criterion of >10⁴ CFU/mL, we likely underestimated the prevalence of MS in the saliva of both the mothers and infants, compared with using the cutpoint of ≥4.5×10³ CFU/mL. Our estimate of the strength of the association of salivary MS in the mothers and the infants in the breastfed group is therefore also probably conservative.

In addition to the higher threshold for detection of salivary MS, a limitation of our study was the relatively small number (n=104) of mother–infant pairs. This precluded further study of the impact of perinatal variables (e.g., delivery mode). A delayed establishment of S. mutans in the vaginally delivered infants compared with infants delivered by cesarean section has been demonstrated. ¹⁹ Also, within 5 minutes of birth, samples from neonates' skin, oral mucosa, and nasopharyngeal aspirate have been shown to closely resemble their own mother's vaginal microbiota for those vaginally delivered neonates, compared with resembling their mothers' skin surface microbiota found in the neonates delivered by cesarean section. ²⁰

In summary, variables of microbial testing, including S. mutans and/or S. sobrinus, cutoff thresholds for CFUs, and age of infant at sampling, and perinatal variables, including delivery mode and feeding method, continue to challenge and change concepts of development of the oral microbial environment.

Conclusions

Our multi-ethnic study of 104 mother–infant pairs from coastal South Carolina found that 12% (12/104) of full-term infants were MS positive in saliva at 4–6 weeks of age using a cutpoint of >10⁴ CFU/mL for positive MS status (Dentocult SM test). Development of detection methods for salivary MS that are easy to use and permit comparable units of CFUs is needed. Infant oral MS status was not solely predicted by race of mother, by delivery mode, by feeding method, or by mother's salivary MS status. However, when stratified by feeding method, the mother's salivary MS status did significantly predict the infant's salivary MS status in the breastfed group. Clarification of the contribution of perinatal factors related to the predentate infant's oral MS status and better understanding of the factors related to the infant's oral environment are needed to guide education and intervention activities.