I.1. Introduction and Overview of the Study
Purpose: To determine if treatment of pregnant women with periodontitis affects gestational age.
Design: This is a randomized, single-blind, multi-center intervention trial. Women between 13-16 weeks (inclusive) of gestation will be enrolled; half will be randomized to receive non-surgical periodontal therapy during their second trimester of pregnancy and half will receive treatment after delivery. Enrollment is expected to take 18 months and subjects will be followed until delivery.
Sample Size: 816 women; 204 at each of four enrollment centers.
pregnant and between 13-16 weeks of gestation
at least 16 years of age,
have at least 20 natural teeth,
have bleeding on probing (BOP) at 35% or more of all tooth sites,
have at least 1 site on 4 different teeth with pocket depth 4 mm and clinical attachment loss 2 mm
unable to provide informed consent or comply with study protocol,
at medical risk as a result of participation,
have multiple fetuses as diagnosed by ultrasound,
require antibiotic prophylaxis for periodontal procedures.
have extensive decay or multiple broken teeth that will most likely result in the subject having less than 20 natural teeth after essential dental care.
Oral and written informed consent will be obtained from eligible women before enrollment. Consenting participants will receive a detailed periodontal examination and will be randomly assigned to receive non-surgical periodontal treatment either during their second trimester of pregnancy or after delivery. Venous blood and subgingival dental plaque samples will be obtained at baseline. All participants will be referred to a general dentist for treatment of carious, fractured, and abscessed teeth as needed. Test subjects will receive scaling and root planning at the enrollment sites prior to 20 weeks of gestation.
All women will be evaluated at monthly intervals until they deliver. Test subjects will receive monthly tooth polishings and oral hygiene instructions; control subjects will receive brief oral examinations only. All subjects will receive full-mouth periodontal examinations at 21-24 weeks and again at 29-32 weeks of gestation. Venous blood and subgingival dental plaque samples will be retaken at 29-32 weeks. An obstetrical nurse will abstract pertinent maternal health and birth information from the subject’s medical record at baseline and at the time of delivery. Control subjects will receive the same periodontal treatment after delivery.
Primary Obstetrical Outcome
Secondary Obstetrical Outcome
Primary Immunological, Microbiological and Clinical Outcomes
Serum C-Reactive Protein
Sum of Serum IgG antibodies to Bacteroides forsythus, Treponema denticola, Porphyromonas gingivalis, Campylobacter rectus, Fusobacterium nucleatum, Prevotella intermedia, and Actinobacillus actinomycetemcomitans
Sum of subgingival counts of B. forsythus, T. denticola, P. gingivalis, C. rectus, F. nucleatum, P. intermedia, and A. actinomycetemcomitans
Bleeding on probing
Secondary Immunological, Microbiological and Clinical Outcomes
Serum IL-1, IL-6, PGE2, TNF, and MMP-9
Specific serum antibodies to B. forsythus, T. denticola, P. gingivalis, C. rectus, F. nucleatum, P. intermedia, and A. actinomycetemcomitans
Individual counts of B. forsythus, T. denticola, P. gingivalis, C. rectus, F. nucleatum, P. intermedia, and A. actinomycetemcomitans
Serum endotoxin level
Clinical attachment level
I.1.2. Background and Rationale
According to the U.S. Surgeon General's 2000 Report on Oral Health, preterm birth and low birth weight are the leading perinatal problems in the United States . (Note: These numbers refer to references given in Section I.1.4 "References for Introduction and Overview".) Despite efforts to improve diagnosis and prevention of preterm birth, it remains a foremost cause of neonatal morbidity and mortality . Established risk factors for preterm birth and low birth weight include tobacco use, drug and alcohol use, hypertension, diabetes, maternal age, low socioeconomic status, African-American ancestry, inadequate prenatal care, poor nutritional status, genitourinary tract infections, multiple pregnancies and short intervals between pregnancies.  Human case control [3, 4], prospective cohort , and intervention studies [6, 7] have also reported an association of preterm delivery or low birth weight with periodontal disease. Animal and bacteriologic studies support this association [8-10].
Preterm birth (< 37 weeks of gestation) occurs in 11.6% of all U.S. births and has risen 9% since 1989-90 and 23% since 1981 . The preterm birth rate varies considerably among ethnic groups. In 1998, the preterm singleton birth rate was 16.15% for non-Hispanic blacks, 11.4% for Hispanics, and 8.59% for non-Hispanic whites. Very preterm birth (< 32 weeks of gestation) occurs in 1.96% of U.S. births and mild to moderate preterm birth (32-36 weeks) occurs in 9.63% of all births. As noted by Ventura et al. , preterm birth, especially very preterm birth, is associated with long-term neurodevelopmental and respiratory disorders and is a major cause of infant mortality [11, 12].
Low birth weight (LBW), defined as less than 2500 grams, occurs in 6.05% of all singleton U.S. births and has remained essentially unchanged since 1989 . The incidence of LBW differs by race and ethnicity. For singletons, it occurs in 11.4% of non-Hispanic black births, 6.8% of American Indian births, 5.3 - 8.2% of Asian and Pacific Islander births, 5.4% of Hispanic births, and 4.9% of non-Hispanic white births. Infants may have LBW because they are born prematurely at a weight appropriate for their gestational age, or because they are small for their gestational age (SGA) due to intrauterine growth restrictions. SGA infants have been defined either as those falling below the 10th percentile of birth weight for their gestational age or as having birth weight more than two standard deviations below the population mean (roughly the 3rd percentile) .
In the Collaborative Perinatal Project, histologic chorioamnionitis was found to be associated with fetal growth restrictions in term and preterm infants . Since intrauterine infection may be associated with poor fetal growth, it is possible that maternal periodontal infection could affect intrauterine fetal growth that leads to SGA infants.
Extremely low birth weight infants (under 1000 grams) are at significant risk for neurologic abnormalities and delays in development and function . In addition to developmental and social problems, medically managing LBW babies is very costly. In the U.S., an estimated $5.4 billion is spent annually managing LBW babies, which is about 10% of the annual health care expenditures for children .
I.1.2.1. Association of Preterm Delivery and Low Birth Weight and Periodontal Disease. Several independent investigators have reported an association between premature birth, low birth weight and moderate to severe periodontal disease [3-5, 7]. The risk for delivering a preterm or low birth weight baby, due to either premature labor or premature rupture of membranes, is increased in women with the poorest periodontal health. The relationship remains after adjustments for known risk factors such as tobacco use, drug use, alcohol consumption, prenatal care, parity, genitourinary infection, and nutrition. It has been estimated that 18% of the risk for preterm low birth weight is attributable to severe periodontal infection.
In a prospective study of 1,313 women, those with severe or generalized periodontal disease, defined as having at least 90 tooth sites with CAL 3 mm, were found to be at elevated risk for experiencing both preterm birth [OR = 4.5 (2.2 - 9.2)] and very preterm (< 32 weeks) birth [OR = 7.1 (1.7 - 27.4)] .
Only two intervention trials have been reported in the literature. While neither is definitive in nature, both found that periodontal treatment decreased the incidence of preterm birth in women with periodontitis  and gingivitis . In a non-randomized study, Mitchell-Lewis et al.  found that scaling and root planing and oral hygiene instructions reduced the risk of preterm low birth weight in teenagers with gingivitis. Preterm birth or low birth weight (PBLBW) occurred in 18.9% of women who did not receive therapy and in 13.5% of those who did (p = 0.36). Although this difference was not statistically significant, it represented a 28.6% reduction in PBLBW in the treatment group. This population had gingivitis, but little if any periodontitis.
In a larger randomized clinical trial of 400 pregnant women with periodontitis, non-surgical treatment delivered before 28 weeks of gestation was associated with a significant reduction in the number of preterm and low birth weight babies . The incidence of preterm birth (< 37 weeks) and low birth weight (< 2500 grams) were 1.10% and 0.55%, respectively, in test subjects, and 6.38% and 3.72% in untreated controls. However, twenty-nine women in the test group (18%) received supplemental antibiotics for severe aggressive periodontitis, which likely altered the course of non-oral infections as well. No control women received similar antibiotic therapy.
I.1.2.2. Animal studies. The relationship between periodontal infections and adverse pregnancy outcomes is supported in animal models as well. In hamsters, inoculations with lipopolysaccharide (LPS) from Escherichia coli and the periodontal pathogen Porphyromonas gingivalis produce a dose-dependent reduction in fetal weight . Pregnant hamsters infected with P. gingivalis also experience significant reductions in fetal litter weight when compared to controls . The effect is seen even at P. gingivalis levels that are insufficient to cause fever or decreased body weight in the dams. Subcutaneous inoculations with periodontal pathogens in hamsters stimulates local production of tumor necrosis factor alpha (TNF-) and prostaglandin E2 (PGE2). Experimentally-induced periodontitis is also accompanied by a significant rise in amniotic PGE2 , which may explain the link between oral infection and changes in the fetal environment .
I.1.2.3. Immunological responses. Bacterial infections frequently stimulate a systemic acute phase response manifested by increased production of some 25 plasma proteins [18-20]. Periodontal and gingival inflammation results in a dose-dependent increase in levels of both local and systemic acute phase proteins, including CRP and haptoglobin [21-27].
Chemokines and proinflammatory and immunomodulatory cytokines are produced by the amniotic and decidual membranes and have been detected in the fetal circulation. Many of these mediators have been associated with preterm delivery. For example, PGE2 concentration is increased in normal physiologic parturition and is positively correlated with preterm birth/fetal growth retardation [28, 29]. Modulation of the PGE2 receptor by infection could raise the risk of preterm labor . However, these prostanoid changes, along with changes in local cytokines, are most frequently associated with intra-amniotic infections [29, 30].
Several inflammatory mediators, including PGE2 , IL-6 [18, 32], and G-CSF may be potential biomarkers for preterm birth. When compared to controls, levels of IL-6 and/or IL-8 are higher in serum of women with premature rupture of membranes (PROM) and amniotic infections. Elevated levels of IL-6 and IL-8 may also predict a shorter interval between PROM, preterm labor, and delivery . Elevated levels of IL-6 were found to predict delivery within 48 hours and preterm delivery before 34 weeks in women without evidence of chorioamnionitis . Levels of IL-6, IL-8, and TNF- in serum correlate with those in amniotic fluid [35-37].
I.1.2.4. Periodontal infection and bacteremias. Simple tooth brushing, flossing, and chewing can produce a bactermia [38, 39]; the likelihood is at least doubled when oral infections, such as periodontal disease, are present . Although periodontal infections are polymicrobial, some species are consistently elevated in diseased sites. The oral microorganism Bacteroides forsythus appears most strongly associated with an increase in risk for preterm birth. Treponema denticola, P. gingivalis, Actinobacillus actinomycetemcomitans, and Campylobacter rectus have also been identified in the subgingival dental plaque of women who experience preterm delivery .
Bacterial vaginosis is an established risk factor for preterm birth. Fusobacterium and Bacteroides species are frequently present in vaginosis cultures and have been associated with an increased risk of LBW. F. nucleatum, when present in the amniotic fluid, is strongly associated with LBW in mothers with intact membranes [40, 41]. Hill suggested that this microorganism might enter the birth canal and amniotic fluid through repeated bacteremias . It can invade oral epithelial cells and modulate cytokine expression , and has the potential to elicit a significant immune response in the mother.
While it was not significantly associated with LBW in the New York study , levels of P. intermedia have been shown to increase during pregnancy .
I.1.2.5. Biologic Rationale and Working Hypothesis. The systemic effect of repeated transient bacteremias in pregnant women is unknown. However, since periodontal disease has been associated with increased risk of preterm delivery, the systemic release of cytokines and/or lipopolysaccharides from gram-negative oral bacteria may affect the placenta and/or fetus, causing preterm labor and/or low birth weight. . The rationale for this clinical trial is that by reducing the incidence of repeated gram-negative bacteremias and the systemic release of lipopolysaccharides or inflammatory cytokines, preterm birth and its associated morbidity and mortality may also be reduced.
The central working hypothesis for this clinical trial is that non-surgical periodontal therapy for pregnant women with periodontitis will reduce the incidence of preterm birth.
I.1.3. Study Design
I.1.3.1. Overview. This multi-center, single blind, randomized, controlled clinical trial is designed to determine if periodontal treatment in pregnant women with periodontitis increases the gestational age of pre-term infants, thereby reducing the incidence of pre-term birth. Eight hundred sixteen women, between 13-16 weeks of pregnancy as determined by menstrual history and obstetrical ultrasound, will be randomized at four sites. After collection of baseline data, subjects at each enrollment site will be randomly assigned to receive either non-surgical periodontal therapy (test subjects) or the same treatment post-partum (control subjects). All subjects will be offered essential dental care, i.e., treatment for abscessed teeth and dental caries diagnosed. All subjects will be seen for monthly obstetrical and dental visits until they deliver.
To maintain periodontal health, test subjects will receive supragingival polishing and oral hygiene instruction at monthly intervals until 36 weeks or delivery. Control subjects will receive brief monthly dental examinations without periodontal treatment as "attention placebo" visits during the same period. Both groups will have periodontal data recorded and subgingival plaque collected for microbial analysis by trained and calibrated examiners at Baseline (13-16 weeks) and again at 21-24 weeks, and 29-32 weeks of pregnancy. Both groups will have serum collected to measure intermediate links in the putative causal chain between the gums and the uterus. The periodontal health of all subjects will be monitored at regular intervals. Any subject who experiences progressive periodontitis during the course of the study will be provided non-surgical treatment.
I.1.3.2. Subject population, study intervention, outcome measures.
Subject Selection and Informed Consent. Potential subjects will be screened at the obstetrics clinics of the respective enrollment centers. Consenting subjects will complete a brief medical history and undergo a brief intra-oral examination to determine likely eligibility. Before a subject is enrolled in the trial, oral and written informed consent will be obtained in accordance with the policies of the respective institution’s Institutional Review Board. Interpreters, for non-English speaking subjects, will be available to assist in the consent process.
Inclusion Criteria. To be eligible for randomization in this study, each subject must:
1. Be pregnant in the first 13-16 weeks of gestation as determined by menstrual history and obstetrical ultrasound, except at the Kentucky site (see Section II.1.1.1.)
2. Be at least 16 years of age,
3. Have at least 20 natural teeth,
4. Have bleeding on probing (BOP) at 35% of all sites, and
5. Have 1 site on 4 different teeth with pocket depth 4 mm and 2 mm
Exclusion Criteria. Subjects will be excluded from participation if they:
1. Are unable to provide informed consent or are unable to cooperate with the study protocol.
2. May be placed at medical risk as a result of participation (i.e. subjects with hematologic disease or other disorders that preclude routine non-surgical periodontal therapy).
3. Have multiple fetuses as diagnosed by ultrasound.
4. Require antibiotic prophylaxis for periodontal procedures
5. Require greater than 3 essential dental care visits because of extensive decay and/or broken teeth and are likely to have less than 20 natural teeth following essential dental care.
Test group intervention: Non-surgical periodontal therapy. Subjects randomized to the test group will undergo one to four 1.5-hour sessions of full mouth scaling and root planing plus intensive oral hygiene instruction. Local or topical anesthesia will be used as needed. This treatment will be completed before 20 weeks of pregnancy. Test subjects will also receive monthly supragingival tooth polishing and reinforcement of oral hygiene until they deliver. The purpose of these interim visits is to maintain plaque control and periodontal health. Such frequent tooth polishings and oral hygiene
instructions have been shown to optimize periodontal health and to maintain the reductions in subgingival periodontal pathogens achieved with scaling and root planning. If there is obvious bleeding at tooth sites due to new or residual calculus, these sites will be re-scaled/root planed at the monthly follow-up appointments.
Control group: Delayed periodontal therapy. Subjects randomized to the control group will receive brief oral exams at their monthly obstetrical follow-up visits. No periodontal treatment, cleanings or oral hygiene instructions will be provided to a subject unless she experiences progressive periodontitis. Control subjects will be asked to refrain from receiving cleanings or periodontal treatment by non-study providers during the trial. Subjects who receive such care will receive no more study therapy and their last available periodontal data will be carried forward on an intent-to-treat basis. After delivery, all control subjects will be offered the same scaling and root planing and oral hygiene instructions provided to test subjects.
The purpose of the control group’s monthly exams is to provide test and control groups with the same number of dental visits. Without controlling for number of dental visits, test subjects might benefit from a differential placebo effect.
Outcome measures. The outcome measures are as follows.