Abstract
Background
Mexican-American children have a higher caries prevalence than the US average. The Mothers and Youth Access (MAYA) study was a randomized clinical trial initiated to address this problem.
Aim
Comparison of the efficacy of two prevention interventions in reducing early childhood caries (ECC).
Design
All 361 randomized mother-child dyads received oral health counseling. Beginning at 4 months postpartum, intervention mothers received chlorhexidine (CHX) mouthrinse for 3 months beginning 4 months postpartum and children received fluoride varnish (FV) every 6 months from age 12–36 months. Control group children received FV if precavitated lesions developed. Salivary mutans streptococci (MS) and lactobacilli were assessed.
Results
No significant difference in children’s 36-month caries incidence between groups; 34% in each group developed caries ((d2+fs) > 0). About half of control group developed precavitated lesions and received therapeutic FV. Maternal MS levels declined during CHX use, but increased when discontinued.
Conclusions
Maternal postpartum CHX regimen, oral health counseling, and preventive child FV applications were not more efficacious than maternal counseling with child therapeutic FV for precavitated lesions for ECC prevention. FV for young children with brief maternal CHX use and oral health counseling may need to be combined with additional or longer-term therapies to significantly reduce ECC in high-risk populations.
Keywords: Epidemiology, Fluoride, Perinatal, Oral Health, Infant, Early Childhood Caries, Randomized Clinical Trial, Chlorhexidine, Prevention
Introduction
Early Childhood Caries (ECC) remains the most prevalent chronic US childhood disease, 1,2 increasing from 24% in 1988–1994 to 28% in 1999–2004.3 ECC poses a serious threat to child welfare, particularly among economically disadvantaged, underserved children4,5 with sequelae of pain, difficulty eating, speech impediment and being underweight.3 Mexican-American children have ECC at 3.5–4.6 times greater prevalence than the US population.6 ECC’s adverse health effects, economic costs, and racial/ethnic and income disparities call for improved prevention strategies. Dental caries can begin in 10–12 month old children.7–9 Research showed caries in 2.5 to 3.8 year olds may progress rapidly in two years.10–12 Although ECC is a chronic, multifactorial, transmissible, infectious disease, it is largely preventable. When the Mothers and Youth Access (MAYA) study began, ECC prevention guidelines were not evidence-based. Ismail reviewed 130 papers from 1966–1997 concluding, “No well-conducted clinical trial on the prevention of ECC was identified.”13 Caries’ biologic mechanisms, fairly well understood in adults, may differ in young children with newly erupted teeth, vertical and horizontal bacteria transmission, fragile enamel, and immature immune systems.14 Chlorhexidine (CHX) mouth rinses are broad spectrum antimicrobials that can reduce plaque, caries and gingivitis.15 It was hypothesized that reducing maternal salivary MS levels during a child’s tooth emergence could delay or prevent MS colonizing their primary dentition, thereby decreasing caries incidence.16 Fluoride varnish (FV) is a high-concentration topical fluoride (22,500 ppm) applied to teeth with a small brush that prevents demineralization and promotes remineralization of early enamel lesions.
At the San Ysidro Health Center (SYHC) in San Diego County, California, 69% of patients live below the federal poverty level; about 95% are Hispanic. In 2000, SYHC 2–4 year olds had 68% caries prevalence (dmfs of 7.9)17 compared to 28% national prevalence (dfs of 2.4) in 2–5 year olds in 1999–2002.18 The MAYA study was developed specifically for this community, which has high ECC prevalence, low socioeconomic status, limited dental care access, and no optimally fluoridated municipal water. Three interventions were selected: 1) CHX mouth rinse to reduce mothers’ cariogenic bacteria and subsequent bacterial transmission to children, 2) FV to strengthen and remineralize child’s tooth enamel, and 3) parental oral health counseling to encourage positive oral health behaviors.
In this dental examiner masked, randomized clinical trial for a population of primarily underserved Mexican-American pregnant women and their newborns, the objective was to compare the efficacy of two strategies in reducing caries incidence and increment; (1) the Control Group receiving (a) parental oral health counseling and (b) therapeutic (“rescue”) FV (TherFV) in children developing precavitated lesions; and (2) the Intervention Group receiving (a) parental oral health counseling, (b) a three-month regimen of CHX mouth rinse for new mothers, and (c) preventive FV applications for children every six months from 12 to 36 months of age. We hypothesized that the more intensive intervention strategy geared at both reducing caries risk factors and increasing protective factors would be more efficacious than the control group’s more limited intervention.
Methods
MAYA was a collaboration involving the University of California, San Francisco (UCSF) and Los Angeles (UCLA), San Diego State University (SDSU), and San Ysidro Community Health Center (SYHC), with approval by three University Institutional Review Boards.
Study Location and Participants
The study was conducted at SYHC, a federally qualified health center near the US–Mexico border and one of the world’s busiest land border crossings.19 Women were eligible if they were: age 18–33; in the second trimester of a normal, single fetus pregnancy; resided in the South San Diego Bay area; supplied evidence of geographic stability; were registered patients at SYHC; and provided signed informed consent in English or Spanish. Women who were recent immigrants, who intended to return to Mexico to deliver their babies, who had high-risk pregnancies, or who had previous medical conditions, pregnancy complications, or pregnancy-related issues that required hospitalization were all excluded from study participation.
Recruitment
Given the highly transitory population, considerable efforts focused on recruitment and retention in three areas: (1) barrier reduction, (2) incentives, and (3) relationship building. The Hispanic bicultural and bilingual principal investigator (PI), a US citizen with extensive research experience, led the team to develop culturally sensitive research materials. To gain acceptance and support, community members, including longstanding health care providers, the clinic board, local organization representatives, regional policy makers, and local residents recommended recruitment strategies and hiring outreach coordinators. Further recruitment and retention details and success are described elsewhere.20 Recruitment occurred between 2003–04 and randomization between 2004–06. The 36-month outcome assessment occurred between 2006–08.
Randomization and Masking
Mothers and their infants were randomly assigned (using computer-generated permuted blocks with varying block sizes securely stored in the web-based data management system-concealing group until assignment) to one of two treatment groups at 4 months postpartum. The timing of randomization was based on the Data and Safety Monitoring Board (DSMB) suggestion that randomization be moved from enrollment during the second trimester to 4 months post-partum to follow the principle of having randomization be as close to the initial intervention as possible. Thus, the period between enrollment and randomization would be akin to a run-in period and would reduce the post-randomization loss to follow-up. The MAYA dental examiners were masked to randomization and did not see participant charts. Separate physical locations for each activity aided examiner masking. Unblinded research team members counseled mothers and distributed CHX rinse. If the CHX caused enamel staining, the mother’s teeth were cleaned prior to being seen by the examiner.
Data Collection and Intervention Protocols
Study procedures for each mother-child dyad included multiple components: parent questionnaire, dental examination, saliva collection for bacterial levels, oral health counseling, monitoring for adverse events, referral of mother and/or child for dental treatment if signs of dental caries were identified, and chart review for possible FV applications outside the study (Illustration 1). In addition, dyads in the intervention group received CHX mouth rinse for the mother (Peridex™ chlorhexidine gluconate 0.12% oral rinse, 3M ESPE OMNII Oral Pharmaceuticals, St Paul, MN) and preventive FV for the child (CavityShield™ 5% NaF varnish, 3M ESPE OMNII Oral Pharmaceuticals, St Paul, MN).
Illustration 1.
MAY A Flowchart
Parental Questionnaires
Trained bilingual, bicultural MAYA staff, with strong community ties, administered questionnaires assessing mother and child oral health-related factors, in English or Spanish, with the mothers at each visit. Data were entered directly into a custom, secure, web-based data management system with built-in validation rules. Questionnaire information included socio-demographics, medical, biological, and behavioral factors potentially related to caries. At each visit, mothers were asked if their children received FV elsewhere. If the answer was yes (there were 6 children receiving non-MAYA FV), they were advised to only obtain FV at MAYA visits. SYHC teams providing no-cost FV to children at WIC or health fairs asked parents if they were MAYA participants; if yes, they did not receive FV.
Dental Exams, Chart Review and Examiner Reliability
The study’s PI (FRG) trained and calibrated two MAYA dental examiners. Following universal infection control, examiners, masked to treatment group assignments, assessed mother and child caries status. NIDCR caries diagnostic criteria were used21 supplemented with diagnostic criteria for non-cavitated (d1) lesions.22 A dental operatory was used. Mothers received an American Dental Association Council on Dental Health Type 3 Inspection with a mouth mirror, explorer and illumination. Child intra-/extra-oral exams were conducted in the knee-to-knee position.23
Replicate exams were performed on dyads throughout the study for quality assurance (QA); 24–50 children participated per QA assessment. Inter-examiner caries reliability kappa statistics for any two examiners ranged from 0.58 to 0.95 and from 0.71 to 0.87 for intra-examiner reliability. Participants had “reversals” (true remineralization or examiner variability) on 410 surfaces recorded initially as carious or demineralized and then sound at a subsequent visit. Of the 26,540 tooth surfaces, 310 demineralized surfaces (1.1%) were subsequently declared as sound; 19 cavitated carious surfaces (0.1%) declared as demineralized surfaces; and 81 cavitated carious surfaces (0.3%) declared sound; 2 (<0.1%) filled/restored tooth surfaces (in the same child) declared sound. Random audits revealed no systematic, consistent or frequent reversal pattern.
Saliva Collection and Microbial Analysis
To assess changes in dyads’ mutans streptococci levels (S. mutans and S. sobrinus) and lactobacilli levels, saliva samples of 1.5–2.0 ml were collected from all dyads. Mothers provided saliva samples at enrollment (about 3 months prior to giving birth), and with their children at 4, 9, 12, 24 and 36 months postpartum. Mothers expectorated saliva into a collection tube; children’s saliva was obtained via a small plastic dropper or “spitting” technique. For QA and reliability, split samples, one with the participant ID and one with a non-participant number, were both analyzed masked to determine reliability (intraclass correlations from 0.60 to 0.98 for MS levels). Saliva remained refrigerated until transfer to UCSD Hospital Microbiology Laboratory, where within 24 hours, serial dilutions and plating were performed to culture MS and LB.24 Following 48–72 h anaerobic incubation, LB colonies were enumerated and MS plates were refrigerated and shipped every other week to UCSF for enumeration.
Interventions
Beginning at 4 months postpartum, intervention group mothers were instructed to rinse twice daily for three months with 0.5 oz. CHX on a 14-day rinse, 14-day rinse-free interval25 based on the hypothesis that CHX would be most efficacious in reducing maternal bacteria levels and transmission during the initial eruption stages of the child’s primary dentition. Mothers were counseled to maintain daily oral hygiene. Intervention group children received 0.25 mL FV containing 5.6 mg fluoride every 6 months, from 12 to 36 months.26–27 Prior to treatment, as part of the informed consent process, participants were advised of possible CHX and FV side effects. If a participant reported any CHX-related adverse event (AE), immediate stoppage was advised. All AEs were recorded and a MAYA dental provider evaluated the participant within a week. Once a CHX-related AE was confirmed, discontinuation was permanent. If the AE was unrelated to CHX, a two-week stoppage with resumption after symptoms dissipated was recommended.
At each visit, mothers received one-on-one oral health counseling in Spanish or English. Topics included were specific to the age-appropriate American Academy of Pediatric Dentistry (AAPD) recommendations for anticipatory guidance in pediatric dental care4,28 and included oral hygiene, diet and nutrition, appropriate fluoride use, pacifier use, ECC, nursing bottle use, weaning at appropriate times, sugar’s role in dental caries initiation, and injury prevention. Culturally suitable scripts were developed and tested before implementation. Four bilingual, age-specific educational brochures (prenatal, infants, toddlers, and 2–5 years) reinforced oral health for the mothers and their children. During the clinical exam, the examiner instructed mothers on brushing children’s teeth and tongue.
Control group children who developed precavitated “white spot” lesions (d1), immediately received therapeutic FV (TherFV) every 6 months as a “rescue” treatment for the study duration. Precavitated lesions for intervention group children were recorded and subsequently monitored. Children with cavitated lesions (d2+) received subsequent FV treatment every 6 months and were referred to the SYHC Pediatric Dental Clinic for treatment at discounted rates. The MAYA research assistant or case manager ensured SYHC dental appointments were made, but families were neither required to obtain treatment, nor barred from continuing in MAYA. Mothers’ precavitated or cavitated lesions were noted in their charts and they were referred to SYHC for treatment, where discounted services were offered during pregnancy through 10 months postpartum. Participants, who met the financial criteria, continued to qualify for discounted dental services.
Sample Size
Based on SYHC’s 2000 data,15 sample size was calculated for difference in dmft between intervention and control groups with a two-sided (alpha=5%) t-test to provide 90% power, under various scenarios using a conservative age-specific standard deviation (5.37), mean dmft of the oldest age group for the control group (5.85) at age 3, and intervention group estimate of 3.5 at age 3, for 98 children per group completing the study. 1998 SYHC reports gave attrition estimates over time in this mobile, high risk population, so enrollment sample size was increased to 239 per group (478 total). The Data and Safety Monitoring Board (DSMB) suggested randomizing at 4-months postpartum instead of during pregnancy corresponding to total enrollment of 512 and later recommended a “rescue” therapeutic FV (TherFV) for ethical reasons on any control group participants with precavitated lesions, which would reduce effect size and required increasing total enrollment to 550, expected randomization total to 334 mother-child dyads, and those completing 36-month follow-up to 226.
Statistical Analyses
Descriptive statistics summarized baseline factors. A multivariate test assessed baseline treatment group imbalance to avoid inflating alpha (Type I error). Intention-to-treat was used for primary analyses. Supplemental analyses utilized participants “as treated”. Primary analysis compared 3-year children’s caries increment (d2+fs) between control and intervention groups with a negative binomial generalized linear model. Three-year caries incidence (d2+fs>0) was assessed between groups with a logit model. As treated analyses used four groups to compare like with like, splitting each of the two randomized groups into whether they had a precavitated lesion and a subsequent FV application. 3-year increment used a negative binomial generalized linear model with a contrast averaging the control-intervention difference for those without precavitated lesions and that for those with precavitated lesions. 3-year incidence used a Mantel-Haenszel test stratifying on precavitated lesion status. Change in mother’s log10 mutans streptococci (MS) level from 4-months postpartum to 9-months postpartum was compared between control and intervention groups with a 2-sample t-test to examine CHX use/compliance in the intervention group. The unmasked data analyst (WS) performed one DSMB interim analysis (Lan-DeMets alpha spending function and O’Brien-Flemming rule) about halfway to study completion with alpha=0.003; thus, alpha=0.0047 applied to these final analyses.
Results
MAYA mothers’ mean age at baseline was 29 years. Most were Spanish speaking, born in Mexico, who had at least a high school education, an annual household income less than $15,000, belonged to a dual parent household, and were recruited from within SYHC. Baseline variables were similar in both randomized groups (10 df chi-square P=0.775). (Table 1) Of the 556 initially pregnant enrollees, 361 (65%) were randomized at the 4-month postpartum visit. Exclusions prior to the 4-month visit were due to missed appointments or relocation (78%). Thereafter, the reasons were high-risk pregnancy (7%), voluntary exit (7%), abortion or child death (2%), and other (6%). Postpartum interval retention was 86% through 9 months, 97% through 12 months, 92% through 18 months, 94% through 24 months, 94% through 30 months, and 99% through 36 months. Overall post-randomization retention, 67.3% (243/361), was almost identical to the planned 67.7%.
Table 1.
Baseline characteristics by randomized treatment group
| Characteristic | Counseling Only* (n=182) | Counseling+CHX+FV (n=179) | Total (N=361) |
|---|---|---|---|
| Mother’s Age (mean ± SD)1 | 28.9±4.5 | 29.1±4.6 | 29.0 ± 4.5 |
| Race/Ethnicity (%, n) | |||
| Mexican/Mexican-American | 93.3 (167) | 94.4 (167) | 93.8 (334) |
| Other Hispanic | 3.4 (6) | 1.1 (2) | 2.2 (8) |
| Other/Non-Hispanic | 3.4 (6) | 4.5 (8) | 3.9 (14) |
| Primary Language at home (%, n) | |||
| Spanish | 60.6 (109) | 62.9 (112) | 61.7 (221) |
| English | 6.7 (12) | 3.9 (7) | 5.3 (19) |
| Both | 32.8 (59) | 31.5 (56) | 32.1 (115) |
| Other | 0.0 (0) | 1.7 (3) | 0.8 (3) |
| Birth Place (%, n) | |||
| Mexico | 57.7 (105) | 63.6 (112) | 60.6 (217) |
| Not Mexico | 42.3 (77) | 36.4 (64) | 39.4 (141) |
| Education Level (%, n)2 | |||
| < 12th Grade | 39.3 (70) | 41.4 (72) | 40.3 (142) |
| ≥12th Grade/vocational | 60.7 (108) | 58.6 (102) | 59.7 (210) |
| Annual Household Income (%, n) | |||
| < $15,000 | 51.8 (88) | 50.9 (88) | 51.3 (176) |
| $15,000–$49,999 | 45.9 (78) | 47.4 (82) | 46.7 (160) |
| >$50,000 | 2.4 (4) | 1.7 (3) | 2.0 (7) |
| Household Composition (%, n) | |||
| Two Parents | 64.1 (116) | 68.0 (121) | 66.0 (237) |
| Single Parent | 35.9 (65) | 32.0 (57) | 34.0 (122) |
includes children who received therapeutic fluoride varnish
Participant ages ranged from 18 to 37. The median age was near the mean. It is possible that younger women may have returned to Mexico for familial support with their young children.
About 60% had at least a high school education (>= HS) or vocational training, but fewer than 10% attended or graduated college. Few of the women had education beyond high school or vocational school.
These data elements were not inclusion/exclusion criteria but the characteristics of the participants.
Outcomes
The change (decline) in mothers’ log10MS from 4 months when the intervention group mothers began CHX to 9 months when saliva was collected and bacteria levels enumerated, and the change during the same time period for mothers in the control group differed significantly by 0.39 (P=0.012; 95% confidence interval of difference in change: 0.09–0.70). This decline provides evidence that the intervention group used CHX rinse during this time and CHX decreased salivary MS levels.(Illustration 2) However, once the CHX was discontinued, maternal MS levels began to steadily increase. By 36 months, 34.2% of the children in each group had cavitated caries incidence (d2+fs>0); most had untreated lesions (d2+s>0). (Table 2) A Fisher’s exact test at 36 months of (d2+fs)>0 between the two groups (P=1.000) and a negative binomial model of d2+fs (P=0.731) were both non-significant. A Mantel-Haenszel test at 36 months of (d2+fs)>0 between the two groups stratifying on d1>0 before 36 months (P=0.817) and a negative binomial model of d2+fs with a contrast of the average difference between the two groups (P=0.808) were also both not statistically significant (Table 3). In the control group, 67 (56%) of 120 children received therapeutic FV (TherFV) subsequently, while 65 (53%) of 123 children in the intervention group had a demineralized lesion before 36 months. Caries incidence (d2+fs>0) compared over time between the groups with a GEE logit model was non-significant (P=0.959) as was increment (d2+fs) with a GEE negative binomial model (P=0.954) (Table 4). Caries incidence (d2+fs>0) compared as an average difference between counseling only and C+CHX+FV groups as well as between C+TherFV and C+CHX+FV+TherFV groups over time with a GEE logit model was non-significant (P=0.782); caries increment (d2+fs) with a GEE negative binomial model was also non-significant (P=0.805) (Table 5).
Illustration 2.
Maternal MS Levels
Table 2.
Intent to treat – Children’s caries incidence (percent (n)) and caries increment (mean (SD)) by 36 months
| Measure | Counseling Only* (n=120) | Counseling +CHX+FV (n=123) |
|---|---|---|
| (d2+fs) > 0 | 34.2 (41) | 34.2 (42) |
| (d2+s) > 0 | 32.5 (39) | 29.3 (36) |
| (d1+fs) > 0 | 54.2 (65) | 55.3 (68) |
| (d1+s) > 0 | 53.3 (64) | 54.5 (67) |
| d2+fs | 2.2 (5.2) | 2.5 (5.5) |
| d2+s | 1.5 (3.2) | 1.4 (3.4) |
| d1+fs | 4.4 (7.0) | 4.5 (6.9) |
| d1+s | 3.7 (5.5) | 3.4 (5.3) |
includes children who received therapeutic fluoride varnishif needed(TherFV)
d1+s includes precavitated & cavitated lesions; d2+s only includes cavitated lesions
Table 3.
As-treated – Children’s caries incidence (percent (n)) and caries increment (mean (SD)) by 36 months
| Measure | Counsel Only (n=53) | Counsel+TherFV (n=67) | C+CHX+FV (n=58) | C+CHX+FV+“TherFV”* (n=65) |
|---|---|---|---|---|
| (d2+fs) > 0 | 13.2 (7) | 50.8 (34) | 8.6 (5) | 56.9 (37) |
| (d2+s) > 0 | 13.2 (7) | 47.8 (32) | 6.9 (4) | 49.2 (32) |
| (d1+fs) > 0 | 26.4 (14) | 76.1 (51) | 29.3 (17) | 78.5 (51) |
| (d1+s) > 0 | 26.4 (14) | 74.6 (50) | 29.3 (17) | 76.9 (50) |
| d2+fs | 0.3 (0.7) | 3.7 (6.6) | 0.3 (1.1) | 4.5 (7.0) |
| d2+s | 0.3 (0.7) | 2.5 (4.0) | 0.2 (0.7) | 2.5 (4.3) |
| d1+fs | 0.6 (1.3) | 7.4 (8.2) | 0.8 (1.7) | 7.8 (8.0) |
| d1+s | 0.6 (1.3) | 6.1 (6.3) | 0.7 (1.3) | 5.8 (6.3) |
C+CHX+FV+“TherFV” are children randomized to C+CHX+FV who would have gotten therapeutic FV at subsequent visit(s) if they had been randomized to the Counsel Only group. [thisis needed to compare like with like; i.e. Counsel+TherFV vs. C+CHX+FV+“TherFV”]
Table 4.
Intent to treat – Children’s caries incidence (percent (n)) by follow-up time
| Follow-up Time (Mo) | (d2+fs) > 0 | (d1+fs) > 0 | ||
|---|---|---|---|---|
| Counsel Only* (n=120) | C+CHX+FV (n=123) | Counsel Only* (n=120) | C+CHX+FV (n=123) | |
| 12 | 0.0 (0) | 0.0 (0) | 6.8 (10) | 8.3 (13) |
| 18 | 5.8(8) | 6.5 (9) | 20.3 (28) | 20.1 (28) |
| 24 | 11.7 (15) | 16.5 (22) | 29.7 (38) | 30.8 (41) |
| 30 | 23.5 (28) | 21.4 (27) | 46.2 (55) | 46.8 (59) |
| 36 | 34.2 (41) | 34.2 (42) | 54.2 (65) | 55.3 (68) |
includes children who received therapeutic fluoride varnish
Sample sizes listed are for 36 months; sample sizes for 12–36 months were respectively,
Counsel only: 146, 138, 128, 119, 120
C+CHX+FV: 155, 139, 133, 126, 123
Table 5.
As-treated – Children’s caries incidence (percent (n)) by follow-up time
| Follow-up Time (Months) |
(d2+fs) > 0 | (d1+fs) > 0 | ||||||
|---|---|---|---|---|---|---|---|---|
| Counsel Only (n=53) |
C+TherFV (n=67) |
C+CHX +FV (n=58) |
C+CHX+FV+ “TherFV” (n=65) |
Counsel Only (n=53) |
C+TherFV (n=67) |
C+CHX +FV (n=58) |
C+CHX+FV+ “TherFV” (n=65) |
|
| 12 | 0.0 (0) | --- | 0.0 (0) | --- | 6.8 (10) | --- | 8.3 (13) | --- |
| 18 | 4.7 (6) | 20.0 (2) | 6.3 (8) | 8.3 (1) | 19.5 (25) | 30.0 (3) | 15.8 (20) | 66.7 (8) |
| 24 | 4.1 (4) | 35.5 (11) | 3.9 (4) | 60.0 (18) | 18.6 (18) | 64.5 (20) | 16.5 (17) | 80.0 (24) |
| 30 | 4.0 (3) | 58.1 (25) | 4.8 (4) | 54.8 (23) | 30.3 (23) | 74.4 (32) | 29.8 (25) | 81.0 (34) |
| 36 | 13.2 (7) | 50.8 (34) | 8.6 (5) | 56.9 (37) | 26.4 (14) | 76.1 (51) | 29.3 (17) | 78.5 (51) |
C+CHX+FV+“TherFV” are children randomized to C+CHX+FV who would have gotten therapeutic FV at subsequent visit(s) if they had been randomized to the Counseling Only group.
Sample sizes in the table are for 36 months; sample sizes for 12–36 months were, respectively,
Counseling only: 146, 128, 97, 76, 53
C+TherFV: 0, 10, 31, 43, 67
C+CHX+FV: 155, 127, 103, 84, 58
C+CHX+FV+“TherFV”: 0, 12, 30, 42, 65
MAYA had 6 serious adverse events (SAEs): 3 fetal mortalities, which were unrelated to study, prior to randomization; 3 intervention group mothers became pregnant post-randomization, discontinued CHX rinse, and had normal healthy deliveries. Moreover, there were 15 occurrences of mothers’ dissatisfaction with dental staining likely associated with CHX; each mother was offered a free SYHC dental cleaning. 917 FV applications were performed with no subject safety issues and no AEs, corresponding to a 1-sided upper 95% confidence interval of a FV application-related AE of 0.33%.29
Discussion
Interpretation
MAYA began with a population at extremely high caries risk. About half the control group developed precavitated lesions and subsequently received therapeutic “rescue” FV (TherFV) for ethical reasons in these very young children. Thus, the “intent-to-treat” analysis includes children in the control group with and without FV, diluting any possible differences between groups. The control group “as treated” who did not develop precavitated lesions is at much lower caries risk. Although outside discounted dental services were available to MAYA mothers and children, mothers had an extremely high prevalence of caries and untreated cavitated lesions at study baseline30. Analysis showed a decrease in intervention group maternal bacteria levels during their 3-month CHX rinse regimen. However once CHX was terminated, bacteria levels increased. Thus, bacterial transmission from mother to child was not likely reduced for this 3-year study period. A longer CHX regimen may be warranted. Horizontal bacterial transmission to child participants from siblings and other children may have also occurred. CHX and/or FV interventions may have given parents a false sense of security differing from the control group. In addition, the lack or unavailability of effective child antibacterial regimens needs future research, since FV alone insufficiently addressed the extremely high baseline risk of these young children. From our previous FV study8 and MAYA, there were no reports of any child’s allergic reaction, sensitivity to FV, or any contraindications with asthmatic patients encountered during the study.
MAYA participants had considerably reduced ECC compared to the same population historically. In 2000, SYHC caries prevalence in 2–4 year olds was 68% (mean dmfs 7.9). In MAYA, at age 3, caries prevalence was 34% (mean d2+fs 2.3 and the m component was very small in this age group). Although there were no differences between MAYA’s control and intervention groups, participants had one-half the ECC prevalence and less than one-third the caries severity compared to their community’s historical data. Well-timed and consistent maternal oral health counseling delivered in a culturally sensitive manner to both groups may have played an important role in reducing caries incidence for both groups. Importantly, our prior study demonstrating significant caries prevention from FV excluded children with caries at baseline when they were 6–44 months old8 while this study included all children from 4 months of age; thus, this study included children at even higher risk than our previous study. Fluoride varnish for children and CHX and counseling for mothers did not provide significantly more caries prevention in this group than counseling alone. Since participants were recruited from a single, US-Mexico border city, with a similar socio-demographic profile, their cultural/social idiosyncrasies (e.g. dietary practices, social structures, and immigration status) may diminish the generalizability.
Overall Evidence
Parental counseling, a 3-month maternal CHX regimen, and FV applied every 6 months to children’s teeth was not shown to be sufficient to significantly reduce caries; and was not more efficacious than parental counseling alone combined with therapeutic fluoride varnish for precavitated lesions. The findings raise additional questions on how to effectively and substantially reduce caries in this extremely high caries risk population. A multi-faceted intervention program may be required that considers the following: a) an age appropriate antibacterial regimen for very young children, b) FV application frequency, c) maternal CHX rinse duration, d) length and type of parental oral health counseling, and e) additional caries preventive strategies based on age and caries risk. Furthermore, the duration and continuum of these interventions may be significant for their cumulative effect on caries prevention.
Bullet Points.
-
What this paper or case report adds:
Parental counseling, a 3-month maternal CHX regimen and FV applied semi-annually to the child’s teeth were not shown sufficient in extremely high caries risk populations to significantly reduce caries and was not more efficacious than parental counseling alone combined with FV alone.
A multi-faceted caries intervention program may not only be required but the duration and continuum of treatment may be significant for their cumulative effect on caries prevention
-
Why this paper or case report is important to paediatric dentists:
Colonization of bacteria occurred at a very early age, prior to the eruption of the first tooth, and continued to increase as the child aged.
Bacterial testing can be considered a good predictor within a disease progression management model and is recommended as part of a comprehensive risk assessment tool.
Salivary MS & LB levels and mother’s age were identified as significant ECC risk factors strongly predicting future caries incidence/increment.
Acknowledgments
The authors acknowledge and express their appreciation to Dr. Charles Hoover and Dr. Lisa Chung of UCSF; Dr Liz Keays of UCSD; Dr. Tracy Finlayson of SDSU; the MAYA staff, for their dedicated work with the MAYA families; Mr. Terry Whitaker, Chief Operating Officer, Dr. Greg Talavera, Dr. Joachim Reimann, Dr. Sergio Cuevas, and the San Ysidro Health Center for providing the site for our research and support for the MAYA staff; all of the families that participated in the MAYA research, and Ms. Debra L. Tom for her editorial assistance.
The study was funded by US DHHS/NIH/NIDCR & NCMHHD U54 DE014251, Center to Address Disparities in Children’s Oral Health; 3M ESPE OMNII Oral Pharmaceuticals provided the chlorhexidine mouth rinse and fluoride varnish for the study.
Footnotes
ClinicalTrials.Gov Identifier: NCT00066950
Declaration of Interests: There are no conflicts of interest for any of the contributing author
References
- 1.Vargas CM, Crall JJ, Schneider DA. Sociodemographic distribution of pediatric dental caries: NHANES III, 1988–1994. J Am Dent Assoc. 1998;129(9):1229–38. doi: 10.14219/jada.archive.1998.0420. [DOI] [PubMed] [Google Scholar]
- 2.Poland C, 3rd, Hale KJ. Providing oral health to the little ones. J Indiana Dent Assoc. 2003;82(4):8–14. [PubMed] [Google Scholar]
- 3.Dye BA, Tan S, Smith V, et al. Trends in oral health status: United States, 1988–1994 and 1999–2004. Vital Health Stat. 2007;11(248):1–92. [PubMed] [Google Scholar]
- 4.Ramos-Gomez F, Crall J, Slayton R, et al. Caries risk assessment appropriate for the age one visit. CDA Journal. 2007;35(10):689–702. [PubMed] [Google Scholar]
- 5.Dental Health Foundation. “Mommy, It Hurts to Chew” The California Smile Survey, an Oral Health Assessment of California’s Kindergarten and 3rd Grade Children. 2006. [Google Scholar]
- 6.Dye BA, Arevalo O, Vargas CM. Trends in paediatric dental caries by poverty status in the United States, 1988–1994 and 1999–2004. Int J Paediatric Dent. 2010;20:132–143. doi: 10.1111/j.1365-263X.2009.01029.x. [DOI] [PubMed] [Google Scholar]
- 7.Psoter WJ, Zhang H, Pendrys DG, et al. Classification of dental caries patterns in the primary dentition: a multidimensional scaling analysis. Community Dent Oral Epidemiol. 2003;31(3):231–8. doi: 10.1034/j.1600-0528.2003.00044.x. [DOI] [PubMed] [Google Scholar]
- 8.Weintraub JA, Ramos-Gomez F, Jue B, et al. Fluoride varnish efficacy in preventing early childhood caries. J Dent Res. 2006;85(2):172–6. doi: 10.1177/154405910608500211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gansky SA, Ramos-Gomez F, Mertz E, et al. California Program on Access to Care findings. California Policy Research Center, University of California, Office of the President; 2007. Fluoride varnish can prevent caries in California’s low-income preschool children. [Google Scholar]
- 10.Thibodeau EA, O’Sullivan DM. Salivary Mutans streptococci and dental caries patterns in pre-school children. Community Dent Oral Epidemiol. 1996;24(3):164–8. doi: 10.1111/j.1600-0528.1996.tb00835.x. [DOI] [PubMed] [Google Scholar]
- 11.Grindefjord M, Dahllof G, Nilsson B, et al. Prediction of dental caries development in one-year old children. Caries Res. 1995;29:343–8. doi: 10.1159/000262090. [DOI] [PubMed] [Google Scholar]
- 12.Grindefjord M, Dahllof G, Nilsson B, et al. Stepwise prediction of dental caries in children up to 3.5 years of age. Caries Res. 1996;30(4):256–66. doi: 10.1159/000262333. [DOI] [PubMed] [Google Scholar]
- 13.Ismail AI. Prevention of early childhood caries. Community Dent Oral Epidemiol. 1998;26(Supp1):49–61. doi: 10.1111/j.1600-0528.1998.tb02094.x. [DOI] [PubMed] [Google Scholar]
- 14.Seow WK. Biological mechanisms in early childhood caries. Community Dent Oral Epidemiol. 1998;26(Suppl 1):8–27. doi: 10.1111/j.1600-0528.1998.tb02090.x. [DOI] [PubMed] [Google Scholar]
- 15.Moran J, Addy M, Wade W, et al. The effect of oxidising mouthrinses compared with chlorhexidine on salivary bacterial counts and plaque regrowth. Clin Periodontol. 1995;22(10):750–5. doi: 10.1111/j.1600-051x.1995.tb00257.x. [DOI] [PubMed] [Google Scholar]
- 16.Tenovuo J, Hakkinen P, Paunio P, et al. The effects of chlorhexidine-fluoride gel treatments in mothers on the establishment of mutans streptococci in primary teeth and the development of dental caries in children. Caries Res. 1992;26(4):275–80. doi: 10.1159/000261451. [DOI] [PubMed] [Google Scholar]
- 17.SYHC dental needs assessment. San Ysidro Health Center; 2001. [Google Scholar]
- 18.Beltrán-Aguilar ED, Barker LK, Canto MT, et al. Surveillance for dental caries, dental sealants, tooth retention, edentulism, and enamel fluorosis --- United States, 1988--1994 and 1999–2002. MMWR Surveill Summ. 2005 Aug 26;54(3):1–43. [PubMed] [Google Scholar]
- 19.Ysidro San. San Diego, California: [Accessed: August 12, 2010]. Available at: http://en.wikipedia.org/wiki/San_Ysidro,_San_Diego,_California. [Google Scholar]
- 20.Ramos-Gomez FJ, Chung LH, Gonzalez-Beristain R, et al. Recruiting and retaining pregnant women from a community health center at the US–Mexico border for the Mothers and Youth Access clinical trial. Clinical Trials. 2008;5:336–346. doi: 10.1177/1740774508093980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.U.S. Department of Health and Human Services, Public Health Services, National Institute of Health, National Institute of Dental Research. NIH Publication No. 91-2870. Bethesda, MD: 1991. Oral health surveys of the National Institute of Dental Research, Diagnostic criteria and procedures. [Google Scholar]
- 22.Drury TF, Horowitz AM, Ismail AI, et al. Diagnosing and reporting early childhood caries for research purposes. A report of a workshop sponsored by the National Institute of Dental and Craniofacial Research, the Health Resources and Services Administration, and the Health Care Financing Administration. J Public Health Dent. 1999;59(3):192–7. doi: 10.1111/j.1752-7325.1999.tb03268.x. [DOI] [PubMed] [Google Scholar]
- 23.Ramos-Gomez F. Clinical considerations for an infant oral health care program. Compend Contin Educ Dent. 2005 May;26(5 Suppl 1):17–23. [PubMed] [Google Scholar]
- 24.Shi W, Jewett A, Hume WR. Rapid and quantitative detection of Streptococcus mutans with species-specific monoclonal antibodies. Hybridoma. 1998;17:365–71. doi: 10.1089/hyb.1998.17.365. [DOI] [PubMed] [Google Scholar]
- 25.Brambilla E, Felloni A, Gagliani M, et al. Caries prevention during pregnancy: results of a 30-month study. J Am Dent Assoc. 1998;129:871–7. doi: 10.14219/jada.archive.1998.0351. [DOI] [PubMed] [Google Scholar]
- 26.Seppa L, Tolonen T. Caries preventive effect of fluoride varnish applications performed two or four times a year. Scand J Dent Res. 1990;98:102–5. doi: 10.1111/j.1600-0722.1990.tb00947.x. [DOI] [PubMed] [Google Scholar]
- 27.Twetman S, Petersson LG, Pakhomov GN. Caries incidence in relation to salivary mutans streptococci and fluoride varnish applications in preschool children from low- and optimal-fluoride areas. Caries Res. 1996;30(5):347–353. doi: 10.1159/000262341. [DOI] [PubMed] [Google Scholar]
- 28.Nowak AJ, Casamassimo PS. Using anticipatory guidance to provide early dental intervention. J Am Dent Assoc. 1995;126:1156–63. doi: 10.14219/jada.archive.1995.0337. [DOI] [PubMed] [Google Scholar]
- 29.Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. JAMA. 1983;249(13):1743–5. [PubMed] [Google Scholar]
- 30.Weintraub JA, Gansky SA, Santo W, et al. Chlorhexidine mouthrinse efficacy and duration in reducing salivary mutans streptococci. J Dent Res. 2010;89(Spec Iss A):848. [Google Scholar]