Obstetrics and Periodontal Therapy (opt) Study Manual of Operations Version 1 March 3, 2003 Brief Table of Contents



Download 0.98 Mb.
Page3/18
Date conversion04.12.2016
Size0.98 Mb.
1   2   3   4   5   6   7   8   9   ...   18
Primary Obstetrical Outcome




  • Gestational age at birth



Secondary Obstetrical Outcome




  • Birth weight



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

  • Gingival Index

  • Bleeding on probing

  • Pocket depth



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.1.3.3. Procedures.
Timeline and Enrollment Schedule. Twelve subjects will be recruited at each site per month for 17 months. Each site will enroll 204 subjects.
Subject Recruitment. A recruitment hygienist will recruit subjects during a pre-natal visit, usually her first prenatal obstetrical visit. To ease recruiting, potential volunteers will be screened by the hygienist/recruiter for inclusion/exclusion criteria at the sites’ obstetrical clinics. All potentially eligible subjects will be given a structured oral presentation of the study and asked to sign a consent form before they are randomized.
Blinding. This trial is single-blinded: the individuals who measure or record obstetrical, periodontal, microbiological and serological outcomes will be blinded to the subject’s group assignment. Specifically, the obstetrical nurse who abstracts obstetrical data from the subject’s medical record, the periodontal examiner who obtains clinical periodontal data, and the laboratory technicians who assay the dental plaque and serum samples will remain blinded. The obstetrical nurse will know the subject’s name as this person must abstract maternal risk factors and birth outcome data from the subject's record. Similarly, the periodontal examiner will dictate data to a dental assistant who will enter it on a form labeled with subject study ID number.
Obstetrical Data Collection. At each site, the treatment-blinded obstetrical nurse will abstract risk factors for premature birth and birth data from the subject’s clinic record.
Determination of gestational age. In all cases, the potential subject must undergo an ultrasound to determine gestational age. In addition, the resident physician may determine gestational age from the potential subject's menstrual history. In some cases, gestational age will not be well-determined without the ultrasound examination. In these cases, the potential subject can be enrolled if in the resident physician's judgment she is under 17 weeks gestation. However, with the exception of the Kentucky site, she cannot be randomized until the results of the ultrasound are available and she is determined to be eligible.
The OPT trial will use the Maternal-Fetal Medicine Units (MFMU, defined by NIH) criteria for determination of gestational age, as follows:


  1. If the LMP is not certain, the ultrasound measurements obtained at the patient’s first ultrasound examination should be used to determine the study gestational age and EDC.

  2. If the LMP date is certain and the ultrasound confirms this gestational age within the number of days specified in the table below, the LMP derived gestational age is used to determine the study gestational age and EDC.

  3. If the ultrasound determined age does not confirm the LMP generated gestational age within the number of days specified in the table below, the ultrasound is used to determine the study gestational age and EDC.

Potential confounding risk factors for preterm delivery and/or intrauterine growth restriction will be recorded and used as covariates in the data analysis. These include:




  • Previous premature births

  • Low maternal weight gain during pregnancy

  • Low maternal weight at enrollment

  • Parity: full term; premature; abortion; living

  • Drug or alcohol consumption

  • Drug or alcohol addiction

  • Low Socioeconomic status

  • Smoking/Tobacco use

  • Anemia

  • Hypertension

  • Maternal infection

  • Fetal congenital anomalies

  • Polyhydramnios

  • Oligohydramnios

  • Medical management

  • Preterm premature rupture of membranes

  • Maternal age

  • Maternal race

The obstetrical nurse will abstract these risk factors from the obstetrical record. Where appropriate, data will be recorded as a continuous measure to express current and cumulative exposure (e.g., number of cigarettes per day during pregnancy and pack years of smoking). Maternal infections, including bacterial vaginosis and sexually transmitted infectious diseases, are prominent risk factors for preterm birth. Maternal infections will be diagnosed through routine clinical and laboratory methods and the results abstracted from subjects’ medical record. Although we will not stratify the randomization based on maternal conditions listed above, it is unlikely that groups will be unbalanced for these important covariates given the sample size.


Infant birth weight and other delivery and postnatal data will be abstracted from obstetrical records made at the time of birth. The Data Coordinating Center will facilitate abstraction of birth data by regularly sending sites a list of subjects likely to deliver soon.
Each enrollment site Study Coordinator will monitor all chart abstractions on a monthly basis for timeliness and accuracy. Also, the obstetrician PI or her/his designee at each site will audit a random 10% sample of obstetrical charts every three months for accuracy of data abstraction.
Clinical Periodontal Examination. Full-mouth periodontal assessments will be made at baseline, at 21-24 weeks (Study Visit 3) and 29-32 weeks (Visit 5) of pregnancy. Trained and calibrated examiners will record the Gingival Index (GI) [44], pocket depth (PD), the distance from the cementoenamel junction to the gingival margin (CEJ·GM) or clinical attachment level (CAL), and bleeding on probing (BOP) from six sites on each tooth (mesiobuccal, mid- or direct buccal, distobuccal, distolingual, mid- or direct lingual, and mesiolingual). If the examiner records CEJ·GM rather than CAL directly, then the examiner or the chairside assistant will calculate and record CAL from the PD and CEJ·GM measures. Plaque [44] and calculus [45] indices will be assessed on the six Ramfjord index teeth for descriptive purposes. PD and CAL will be measured using standardized color-coded probes (UNC –15; Hu-Friedy Mfg. Co., Chicago IL). With well-trained examiners, measurements using manual probes are as reproducible as those obtained using the automated Florida Probe [46, 47]. It is our experience that a manual probe is much less uncomfortable for patients than the Florida Probe.

The clinical measures and indices will be recorded in the following order:

1. Plaque Index on the Ramfjord teeth

2. Gingival Index on all teeth

3. PD then CEJ·GM or CAL on all teeth

4. BOP on all teeth

5. Calculus Index on the Ramfjord teeth

BOP will be assessed after probing the facial/buccal or lingual/palatal surfaces of each quadrant and not after the entire dentition has been probed. It will be scored as present or absent as described [48]. The Calculus Index of Green and Vermillion [45] will be used to score subgingival calculus.


Delivery of Essential Dental Care. Following the baseline visit, all subjects will be referred to a general dentist as needed for treatment of abscessed or carious teeth detected during the baseline examination. One to three visits are planned to provide this care. Essential dental care will be completed before 20 weeks of gestation.
A general dentist will make all essential care treatment decisions. Such treatment is likely to include placement of temporary and/or permanent composite or alloy restorations, tooth extractions, and endodontic therapy. Radiographs will be taken as indicated to facilitate essential dental care. No fixed or permanent removable prosthesis or permanent crowns will be fabricated as part of the study, though temporary removable partial dentures will be fabricated for esthetic or functional purposes to replace teeth extracted during the study. Otherwise, no cosmetic procedures will be provided as part of essential dental care. A letter will be mailed to a subject's dentist of record (if any) informing her/him of the subject’s participation in the trial. It will briefly explain the study and request that all periodontal or dental hygiene treatment be done through the study.
Subjects will be free to seek dental care outside of the study, but the patient will be financially responsible for any care delivered outside of the study. Subjects will be asked to refrain from receiving periodontal treatment or cleanings with their private dentist during the course of their pregnancy. Any subject who receives such care will receive no further care from the study, and their most recent periodontal data will be carried forward on an intent-to-treat basis. All subjects will be followed through delivery. Upon completion of the study, subjects will be informed of the need for additional dental care and will be given a list of clinics where follow-up care is available if they do not have a dentist of record.
Serum Sampling and Analysis. Venous blood samples will be collected at baseline and again at 29-32 weeks of pregnancy. Serum will be separated and immediately frozen at -70° to -80°C. Approximately once a month, samples will be shipped overnight in dry ice to the University of Kentucky for analysis.
Sequential ELISA methods [49] will be used to measure multiple biomarkers in the serum samples. C-reactive protein (CRP) will be quantitated using a capture ELISA [23, 50]. Interleukin(IL)-1ß, IL-6, IL-8, PGE2 and TNF- will be quantitated by ELISA using mouse monoclonal antibodies [51]. MMP-9 will be assayed using a commercial ELISA kit (R&D Quantikine).
Serum IgG antibody levels to seven periodontal bacteria will be quantified using ELISA methods [52, 53]. The antibody levels will be compared to a standard curve and expressed as ng-µg/ml of IgG reactivity. Samples will be assayed in the sequence of least to most prominent antibody level in the sample sera based upon preliminary studies and the minimal detectable dose for the assays. Finally, endotoxin activity will be analyzed using a commercial kit (Endosafe®, Charles River).
After quantifying the constituents of interest, the remaining sera will be archived at -80° C for possible future additional analysis.
Subgingival Plaque Sampling and Analysis. Prior to randomization at baseline, subgingival dental plaque will be collected from each subject from the 4 deepest, most posterior periodontal pockets that bleed on probing. The same sites will be sampled again at 29-32 weeks. Because we are interested in deriving a measure of systemic exposure to periodontal pathogens, and not in characterizing the microbiota at diseased tooth sites, plaque samples from each subject will be pooled for subsequent analyses. Samples will be obtained using separate, standardized, sterile Gracey curets. Samples will be dispersed in phosphate-buffered saline in micro-centrifuge tubes and frozen at –80° C until shipped to the University of Minnesota for analysis. Once at Minnesota, the samples will be kept frozen at -85° C until analyzed.
Total DNA will be extracted and quantified using the Picogreen™ kit (Molecular Probes). Samples will be assayed by quantitative PCR (qPCR) to determine levels of the species listed as primary and secondary microbiological outcomes. PCR assays will be made species-specific by designing primers that anneal to 16S rRNA variable regions found only in each target species. Those assays will be made quantitative using the Amplifluor™ system (Intergen, Gaithersburg, MD) [54]. The lower limit of detection for the qPCR assays is generally about 100 cells.
All samples will be assayed in triplicate. Data from the plate reader will be read directly into a relational electronic database. Average values from the triplicate readings for each sample and species will be used in subsequent analyses.
Coordination of Visits. Study visits will be scheduled to coincide as much as possible with the subject’s routine monthly prenatal visits. To simplify compliance, periodontal care and data and specimen collection will be done in a dental clinic located in the same building as the obstetrics clinic. Except for the HCMC in Minneapolis, subjects will also receive essential dental care in the same hospital-based dental clinic. At the HCMC, subjects will be referred to the University of Minnesota School of Dentistry (2.0 miles away) for this care.
Subject Payment. At each study visit or dental-care visit, subjects will be given a $20 gift certificate for use at a discount store (e.g., Target, Walmart, K-Mart). The gift certificates can be used to purchase items but cannot be redeemed for cash.
Data Transmission. The Study Coordinator is responsible for maintaining a supply of case-report forms, for transmitting forms to the Data Coordinating Center (DCC), and for data quality control at his/her site. Original data forms will be sent to the DCC weekly by courier such as Federal Express.® The Coordinator will keep a photocopy of all forms in a secure, locked file cabinet. Plaque and serum samples will be collected in vials labeled with the subject's study ID and visit number, then forwarded to Dr. Rudney's or Dr. Ebersole's laboratories at regular intervals. Serum and quantitative PCR data will be recorded automatically in Excel files and sent to the DCC by electronic mail.
I.1.3.4. Safety monitoring.
Monitoring for Progressive Periodontal Disease. All subjects will be evaluated clinically every two months for evidence of progressive clinical attachment loss (CAL), defined as an increase from baseline in clinical attachment loss (CAL) greater than 2 mm. The examiner will compare CAL recordings to baseline values and record the location of sites that display progressive CAL, along with the magnitude of change.
If a subject has fewer than 6 cumulative sites with CAL greater than 2 mm, the subject will be referred to the treatment hygienist for localized scaling and root planning, regardless of the subject’s initial treatment assignment. In this manner, the examiner will remain blinded to treatment assignment. Tooth sites requiring such care will be considered to be exited from the study and their periodontal data will be carried forward on an intent-to-treat basis.
If a subject experiences progressive attachment loss from baseline at 6 or more sites, the subject will be considered to have generalized progressive attachment loss and will be exited from her study-assigned therapy. The subject’s records will be forwarded to the consulting periodontist who will be informed of the subject’s group assignment by the study coordinator. The consultant will review clinical records to confirm the diagnosis of generalized progressive periodontitis and either refer the subject to the treatment hygienist for full-mouth scaling and root planning (if originally assigned to the control group) or for additional assessment and care (if in the test group). For test subjects who experience generalized CAL, the consulting periodontist may procure subgingival plaque samples and submit them to a commercial laboratory for microbial analysis and antibiotic sensitivity testing. The consulting periodontist will review the results from the laboratory, select and prescribe an appropriate antibiotic regimen, and arrange to personally re-instrument the subject or refer her to the treatment hygienist for this treatment. This treatment will not be delayed until postpartum.
Based on published reports of disease progression in untreated populations, a conservative estimate is that 2-3% of subjects will experience progressive disease [55].
Subjects will also be monitored at monthly intervals for additional adverse periodontal events including abscess, necrotizing ulcerative gingivitis, pregnancy tumor, and severe or diffuse inflammation.
Obstetrical Safety Monitoring. Obstetrical adverse events will be detected during the subject's regular pre-natal visits and during post-partum chart abstraction. Most if not all of the study's serious adverse events will be obstetrical in nature and will be detected at a regular pre-natal visit or at delivery. Each enrollment site's obstetrical principal investigator (PI) will implement procedures to ensure that all serious adverse events are reported to the Data Coordinating Center (DCC) within 24 hours. Adverse events of lesser severity will generally be detected during the post-natal chart abstraction and will be reported with birth-related data.
Data and Safety Monitoring Board (DSMB). All subjects will be monitored for preterm delivery and other prenatal adverse events that occur during the course of this trial. Periodontal probing and scaling and root planing cause transient bacteremias. Therefore, while unlikely and not supported by data from other trials, periodontal therapy could increase the incidence of pre-term delivery or other adverse events. To minimize this risk, all periodontal therapy and dental care will be done after the first trimester, when fetal formation is complete and after the time when most spontaneous miscarriages occur. Conversely, periodontal therapy may reduce the incidence of preterm birth and this may become evident before all subjects are enrolled or complete the trial.
Accordingly, a Data and Safety Monitoring Board has been established consistent with NIDCR Policies and Procedures. The DSMB is scheduled to meet in Minneapolis, MN prior to start of enrollment and in the fall of years 1, 2 and 3. Interim conference calls may be scheduled at the discretion of the DSMB, the NIDCR, or the study leadership.
At each meeting after enrollment begins, the DSMB will receive a formal report prepared by the Data Coordinating Center (DCC). In general, the report will include data about enrollment and quality of data collection, baseline comparisons of treatment groups, and treatment group comparisons according to each study outcome. Each report will have an open section containing materials suitable for blinded study staff, and a closed section giving outcomes and treatment-group comparisons. The closed section will be seen only by the DSMB, the NIH program officer, and DCC staff. Treatment groups will be coded as A and B, with the identities of groups A and B sent under separate cover.
I.1.3.5. Statistical considerations.
Statistical Analyses. The DCC will analyze study data for all publications and presentations. During enrollment and follow-up, baseline subject characteristics only will be analyzed; no analyses of results by treatment group will be done for publication or presentation before the study’s end.
All analyses will be by intent-to-treat; that is, each subject will be accounted to the group to which she was randomized.
The primary analysis will compare gestational ages in the two groups using time-to-event analysis, where the event is the end of pregnancy and time-to-event is gestational age. The test will be the log-rank test stratified by clinical site. The unusual feature of this analysis is that full-term births (37 weeks) will be treated only as having occurred at 37 weeks or later (i.e., censored at 37 weeks), because we seek and expect no gestational extension beyond 37 weeks; including them in the analysis decreases power. The main secondary analysis of gestational age will adjust for maternal risk factors for preterm birth. This analysis will use the proportional-hazards model, again censoring at 37 weeks gestational age, stratifying by clinical site, and using the maternal risk factors as predictors. A planned variant analysis will test the treatment effect is consistent across study sites (the treatment-by-site interaction).
Although the sample size (see below) allows 30% lost to follow-up, we do not this high of an attrition rate. In case of differential losses, we will do further secondary analyses using methods recently published by Robins and colleagues [56, 57]. Neither approach is computed with standard software; based on long acquaintance with Dr. Robins, we anticipate adapting the programs written for the above papers. Such software is hazardous, so we will also use a simpler, probably less powerful method easily computed with standard software [58], as implemented by McGuigan et al [59]. Specifically, a subject's propensity to be censored before 37 weeks will be modeled by logistic regression on baseline measures and a subject's non-response weight will be the reciprocal of her fitted probability from this logistic regression.
Analyzing quantitative assays of host inflammation, immune response, and bacterial plaque.. Sums are used as outcome measures for immune response and periodontal infection because in similar cases and preliminary data, sums have given more power than the measures that are summed. For each measure, the outcome in the analysis will be change from baseline at each data collection visit. Based on previous experience, the common logarithm of the measure will conform to normal theory much better than the measure itself. Thus, the outcome will be computed from the common logarithm of {measure + 1}, "1" being added to accommodate zero measurements. Test and Control groups will be compared using the methods described below for clinical measures of periodontal health.
Analyzing clinical measures of periodontal disease. For the primary outcome measures GI, BOP and PD, whole-mouth averages or fractions, as appropriate, will be calculated for each subject. Published studies [60, 61] indicate that these summary statistics conform adequately to normal theory. Thus, for each outcome the unadjusted analysis will use a mixed linear model with change from baseline at each follow-up visit as the outcome; with study site, treatment group, and their interaction as between-subject fixed effects; with time and its interactions with treatment and site as within-subject fixed effects; and with the random effect being subject within site and treatment group. The adjusted analysis will add baseline (subject-specific) maternal risk factors as between-subject fixed effects. Clinical attachment loss will also be summarized and analyzed as above; previous experience indicates that it also conforms reasonably to normal theory. Change from baseline is the outcome because previous experience showed that it eliminates a large component of variance between subjects. Each outcome will be analyzed first with only visit, clinical site, treatment, and interactions as predictors, and will be analyzed second by adding the other maternal risk factors for preterm birth to the list of predictors.
If there is substantial dropout or receipt of non-study therapy, generally the same options are available for these outcomes as noted above, except that the continuous outcome variables here allow somewhat greater flexibility [62].
Most analyses will use SAS version 8, with other programs used as needed, mainly S+ and JMP. The DCC will have a utility that creates SAS-analyzable files from the study’s NOMAD databases. This utility documents fields on the analysis files and automatically generates the same labels used in the NOMAD master files; it will be used to create all analysis files.
Sample Size Determination. The primary analysis will compare test and control groups by time to delivery, measured as gestational age, using the log-rank test. The difference to be detected is measured not in terms of event rate, but in terms of clinically meaningful gestational extensions at different gestational ages. Any effect of periodontal treatment will presumably benefit pre-term, but not full-term infants. Moreover, to be clinically meaningful, the gestation extension must be relatively large for infants with the lowest gestational ages; as the gestational age approaches 37 weeks, smaller extensions become worthwhile.

Published survival rates for infants born at various ages were used to derive clinically worthwhile extensions in gestational age. For estimating sample size, these extensions were: at least 5 weeks for infants otherwise born at < 20 weeks; at least three weeks for infants otherwise born at 25 weeks; two weeks for infants otherwise born at 30-35 weeks; and no extension for infants otherwise born at 37+ weeks. The clinically significant effect of periodontal treatment for all gestational ages was determined by extrapolating values at interim time points (Figure 1a). Gestation extension (vertical axis) is given as a function of gestational age without maternal periodontal therapy. The fractions of children born by each gestational age for the test and control groups are presented in Figure 1b. The solid line for the control group is based on pilot data from two enrollment sites (HCMC in Minneapolis MN and Jackson MS). For example, 17% of the children were born by 35 weeks. The dashed line (test group) was derived by applying the detectable difference in Figure 1a to the solid line for the Control group. In this hypothetical treated population, 9.7% of children were born by 35 weeks gestational age. The power of the log-rank test for various sample sizes was computed by simulation using the time-to-event distributions in Figure 1b and censoring at 37 weeks.


Assuming an  (false-positive rate) of 0.05 and allowing 30% lost to follow-up for the primary outcome, the per-group sample sizes for 80%, 85%, and 90% power are 287, 336, and 405, respectively. This study is unlikely to be replicated, so we chose 90% power and allowed 30% lost to follow-up for the primary outcome, giving a sample size of 405 subjects per group, or 810 total. The final sample size was set at 816 subjects to allow equal numbers to be enrolled at each site.
Data Flow and Management. All data from enrollment sites will be recorded on case-report forms which will be sent in weekly batches to the DCC. A logging system, the forms sequence numbers, and shadow databases will be used to track individual forms and data fields through the entire process.
Data from the two laboratories will be e-mailed to the DCC as it is produced, in Excel files having specified formats. These Excel files will be read directly by the DCC's database-management software and incorporated into the study database.
Data Safety and Monitoring Board Reports. Production of DSMB reports is one of the DCC's highest priorities. Data will be shown in tables and graphs to ease detection of patterns, trends, and group differences. In general, the report will include quality-control data (described below), baseline comparisons of treatment groups, and treatment group comparisons according to each study outcome. Each report will have an open section, suitable for viewing by blinded study personnel, and a closed section giving outcomes and treatment-group comparisons. The closed section will be seen only by the DSMB, the NIH program officer, and DCC staff. Treatment groups will be coded as A and B, with the identities of groups A and B sent under separate cover.
Reporting for Quality Management. The principal investigator (PI) and Study Coordinator at each enrollment site have local responsibility for monitoring the trial. The Study Chairman (Dr. Bryan Michalowicz) and the Statistical Study Manager at the Data Coordinating Center will be responsible for external monitoring. The Study Manager will monitor completeness and timeliness of data forms, data transmission, and error correction. Dr. Michalowicz will monitor overall study management, enrollment, and compliance with protocol as specified in the Manual of Procedures. He will provide oversight to assure that quality data are collected, and that adverse events are reported in a timely manner. He will work closely with NIH/NIDCR staff to oversee the trial as specified in NIH/NIDCR Policies and Procedures for Investigator Initiated Clinical Trials.
Each site’s performance according to recruitment, visit attendance, protocol adherence, error rates on forms, timeliness of response to error corrections, weekly forms packages, and birth-data collection will be described in a monthly quality assessment report. These reports will be sent to the Study Coordinators and will be the subject of monthly conference calls among the sites and the DCC.
Randomization. Separate randomization schedules will be prepared by the DCC for each enrollment site. Randomization schedules will be constructed using randomly alternating, permuted blocks of 2 and 4, assuring that approximately equal numbers of test and control assignments will be made at each center.
Following a subject's baseline examination, the Study Coordinator at her site will call the DCC on a toll-free line, and the DCC registrar will use an interactive randomization program to review eligibility criteria with the Study Coordinator. Once eligibility is verified, the registrar will issue a treatment assignment. After the DCC processes a newly randomized subject, it will issue the site a set of ID labels for that subject, which will include the enrollment code.
I.1.3.6. Training.
Training of Study Personnel. Before enrollment begins, Study Coordinators, Recruitment Coordinators, and Obstetrical Data Recorders from each site will go to the Data Coordinating Center (DCC) to be trained by the DCC's Statistical Study Manager in all procedures, including completion of case-report forms.
Periodontal examiners at each enrollment site will be trained and calibrated before enrollment begins. Periodontal examiners at each enrollment site will travel to the University of Minnesota where they will be trained and calibrated in the Oral Health Clinical Research Center. Examiners will be calibrated for both for intra- and inter-examiner error using Dr. Bryan Michalowicz as the “Gold Standard” examiner. The calibration protocol will involve 5 representative subjects being measured twice by each examiner. Examiners will qualify for the study if they achieve the following criteria:
• GI: At least 80% intra- and inter- examiner exact reproducibility plus 95% intra- and inter- examiner reproducibility within ± 1 index unit.

• CAL and PD: 85% and 90% intra-examiner reproducibility within ± 1 mm respectively; 95% intra-examiner reproducibility within ± 2 mm for both measures; 75% inter-examiner agreement for PD within ± 1 mm and 60% inter-examiner agreement for CAL within ± 1 mm.


The kappa statistic will not be used to assess reproducibility of GI scores because it is sensitive to the skewed distribution, which is common for this index.
Determining bleeding on probing (BOP) is an invasive procedure and sites are more likely to bleed at subsequent passes in a calibration trial. Thus, adequate calibration procedures for BOP do not exist. Instead, examiners will observe each other during training sessions so they may discuss and evaluate the criteria for BOP. In our experience, this helps standardize examiners. The Plaque Index[44] cannot be calibrated because plaque is removed when this index is scored; training in assessing the plaque index will be handled in the same manner as BOP.
Intra-examiner reproducibility will continually be assessed. Examiners will re-measure a randomly selected dental quadrant in 5% of subject visits throughout the duration of the trial. The examiner will have no knowledge beforehand which patients or quadrants will be selected for re-measurement. In the event that examiners do not continue to meet baseline calibration standards, they will be re-trained by the gold standard examiner until these standards are re-met.
I.1.4. References for Introduction and Overview.
1. Ventura, S.J., et al., Births: final data for 1998. Natl Vital Stat Rep, 2000. 48(3): p. 1-100.

2. Lu, G.C. and R.L. Goldenberg, Current concepts on the pathogenesis and markers of preterm births. Clin Perinatol, 2000. 27(2): p. 263-83.

3. Offenbacher, S., et al., Periodontal infection as a possible risk factor for preterm low birth weight. J Periodontol, 1996. 67(10 Suppl): p. 1103-13.

4. Dasanayake, A.P., Poor periodontal health of the pregnant woman as a risk factor for low birth weight. Ann Periodontol, 1998. 3(1): p. 206-12.

5. Jeffcoat, M.K., et al., Periodontal infection and preterm birth: results of a prospective study. J Am Dent Assoc, 2001. 132(7): p. 875-80.

6. Mitchell-Lewis, D., et al., Periodontal infections and pre-term birth: early findings from a cohort of young minority women in New York. Eur J Oral Sci, 2001. 109(1): p. 34-9.

7. Lopez, N.J., P.C. Smith, and J. Gutierrez, Periodontal therapy may reduce the risk of preterm low birth weight in women with periodontal disease: a randomized controlled trial. J Periodontol, 2002. 73(8): p. 911-24.

8. Collins, J.G., et al., Effects of Escherichia coli and Porphyromonas gingivalis lipopolysaccharide on pregnancy outcome in the golden hamster. Infect Immun, 1994. 62(10): p. 4652-5.

9. Collins, J.G., et al., Effects of a Porphyromonas gingivalis infection on inflammatory mediator response and pregnancy outcome in hamsters. Infect Immun, 1994. 62(10): p. 4356-61.

10. Offenbacher, S., et al., Potential pathogenic mechanisms of periodontitis associated pregnancy complications. Ann Periodontol, 1998. 3(1): p. 233-50.

11. Berkowitz, G.S. and E. Papiernik, Epidemiology of preterm birth. Epidemiol Rev, 1993. 15(2): p. 414-43.

12. MacDorman, M.F. and J.O. Atkinson, Infant mortality statistics from the 1997 period linked birth/infant death data set. Natl Vital Stat Rep, 1999. 47(23): p. 1-23.

13. Charlton, V., The Small for Gestational Age Infant, in Rudolph's Pediatrics, H.J. Rudolph AM, Rudolphh CD, Editor. 1991, Appleton & Lange: Nrowalk, CT, San Mateo, CA.

14. Williams MC, O.B.W., Nelson RN, Spellacy WN, Histologic chorioamnionitis is associated with fetal growth restriction in term and preterm infants. Am J Obstet Gynecol, 2000. 183: p. 1094-9.

15. Vohr, B.R., et al., Neurodevelopmental and functional outcomes of extremely low birth weight infants in the National Institute of Child Health and Human Development Neonatal Research Network, 1993-1994. Pediatrics, 2000. 105(6): p. 1216-26.

16. Low Birth Weight in Minority Populations PA-99-045. 1999, Bethesda, MD: National Institutes of Health.

17. Oral health in America: a report of the Surgeon General. J Calif Dent Assoc, 2000. 28(9): p. 685-95.

18. Glibetic MD, B.H., The effect of chronic inflammation on the expression of murine acute phase plasma proteins. Protides of the Biological Fluids, ed. P. H. 1986, New York: Pergamon Press.

19. Steel, D.M. and A.S. Whitehead, The major acute phase reactants: C-reactive protein, serum amyloid P component and serum amyloid A protein. Immunol Today, 1994. 15(2): p. 81-8.

20. Trautwein, C., K. Boker, and M.P. Manns, Hepatocyte and immune system: acute phase reaction as a contribution to early defence mechanisms. Gut, 1994. 35(9): p. 1163-6.

21. Adonogianaki, E., J. Mooney, and D.F. Kinane, The ability of gingival crevicular fluid acute phase proteins to distinguish healthy, gingivitis and periodontitis sites. J Clin Periodontol, 1992. 19(2): p. 98-102.

22. Adonogianaki, E., et al., Acute-phase proteins in gingival crevicular fluid during experimentally induced gingivitis. J Periodontal Res, 1994. 29(3): p. 196-202.

23. Ebersole, J.L., et al., Systemic acute-phase reactants, C-reactive protein and haptoglobin, in adult periodontitis. Clin Exp Immunol, 1997. 107(2): p. 347-52.

24. Leibur, E., et al., Prostaglandin E2 levels in blood plasma and in crevicular fluid of advanced periodontitis patients before and after surgical therapy. Oral Dis, 1999. 5(3): p. 223-8.

25. Noack, B., et al., Periodontal infections contribute to elevated systemic C-reactive protein level. J Periodontol, 2001. 72(9): p. 1221-7.

26. Ranney, R.R., Immunologic mechanisms of pathogenesis in periodontal diseases: an assessment. J Periodontal Res, 1991. 26(3 Pt 2): p. 243-54.

27. Sibraa, P.D., et al., Acute-phase protein detection and quantification in gingival crevicular fluid by direct and indirect immunodot. J Clin Periodontol, 1991. 18(2): p. 101-6.

28. Spaziani, E.P., et al., Modulation of the prostaglandin E receptor: a possible mechanism for infection-induced preterm labor. Obstet Gynecol, 1999. 93(1): p. 84-8.

29. Hsu, C.D., et al., Dual roles of amniotic fluid nitric oxide and prostaglandin E2 in preterm labor with intra-amniotic infection. Am J Perinatol, 1998. 15(12): p. 683-7.

30. Otsuki, K., et al., Lactoferrin and interleukin-6 interaction in amniotic infection. Adv Exp Med Biol, 1998. 443: p. 267-71.

31. Challis, J.R., Molecular aspects of preterm labor. Bull Mem Acad R Med Belg, 1998. 153(5-6): p. 263-70.

32. Turhan, N.O., A. Karabulut, and B. Adam, Maternal serum interleukin 6 levels in preterm labor: prediction of admission-to-delivery interval. J Perinat Med, 2000. 28(2): p. 133-9.

33. von Minckwitz, G., et al., Predictive value of serum interleukin-6 and -8 levels in preterm labor or rupture of the membranes. Acta Obstet Gynecol Scand, 2000. 79(8): p. 667-72.

34. Athayde, N., et al., A role for the novel cytokine RANTES in pregnancy and parturition. Am J Obstet Gynecol, 1999. 181(4): p. 989-94.

35. El-Bastawissi, A.Y., et al., Amniotic fluid interleukin-6 and preterm delivery: a review. Obstet Gynecol, 2000. 95(6 Pt 2): p. 1056-64.

36. Figueroa-Damian, R., J.L. Arredondo-Garcia, and J. Mancilla-Ramirez, Amniotic fluid interleukin-6 and the risk of early-onset sepsis among preterm infants. Arch Med Res, 1999. 30(3): p. 198-202.

37. Keelan, J.A., et al., Cytokine abundance in placental tissues: evidence of inflammatory activation in gestational membranes with term and preterm parturition. Am J Obstet Gynecol, 1999. 181(6): p. 1530-6.

38. Guntheroth, W.G., How important are dental procedures as a cause of infective endocarditis? Am J Cardiol, 1984. 54(7): p. 797-801.

39. Roberts, G.J., Dentists are innocent! "Everyday" bacteremia is the real culprit: a review and assessment of the evidence that dental surgical procedures are a principal cause of bacterial endocarditis in children. Pediatr Cardiol, 1999. 20(5): p. 317-25.

40. Hill, G.B., Preterm birth: associations with genital and possibly oral microflora. Ann Periodontol, 1998. 3(1): p. 222-32.

41. Hillier, S.L., et al., The role of bacterial vaginosis and vaginal bacteria in amniotic fluid infection in women in preterm labor with intact fetal membranes. Clin Infect Dis, 1995. 20 Suppl 2: p. S276-8.

42. Krisanaprakornkit, S., et al., Inducible expression of human beta-defensin 2 by Fusobacterium nucleatum in oral epithelial cells: multiple signaling pathways and role of commensal bacteria in innate immunity and the epithelial barrier. Infect Immun, 2000. 68(5): p. 2907-15.

43. Raber-Durlacher, J.E., et al., Experimental gingivitis during pregnancy and post-partum: clinical, endocrinological, and microbiological aspects. J Clin Periodontol, 1994. 21(8): p. 549-58.

44. Loe, H., The Gingival Index, the Plaque Index and the Retention Index Systems. J Periodontol, 1967. 38(6): p. Suppl:610-6.

45. Greene JC, V.J., The simplified oral hygiene index. J Am Dent Assn, 1964. 68: p. 7-13.

46. Osborn, J., et al., Comparison of measurement variability using a standard and constant force periodontal probe. J Periodontol, 1990. 61(8): p. 497-503.

47. Osborn, J.B., et al., Comparison of measurement variability in subjects with moderate periodontitis using a conventional and constant force periodontal probe. J Periodontol, 1992. 63(4): p. 283-9.

48. Pihlstrom, B., Issues in the evaluation of clinical trials of periodontitis: a clinical perspective. J Periodontal Res, 1992. 27(4 Pt 2): p. 433-41.

49. Steffen, M.J. and J.L. Ebersole, Sequential ELISA for cytokine levels in limited volumes of biological fluids. Biotechniques, 1996. 21(3): p. 504-9.

50. Ebersole, J.L., et al., Systemic manifestations of periodontitis in the non-human primate. J Periodontal Res, 1999. 34(7): p. 358-62.

51. Ebersole, J.L. and D. Cappelli, Gingival crevicular fluid antibody to Actinobacillus actinomycetemcomitans in periodontal disease. Oral Microbiol Immunol, 1994. 9(6): p. 335-44.

52. Giardino, A., J.L. Ebersole, and S.C. Holt, Characteristics of systemic antibody responses of nonhuman primates following active immunization with Porphyromonas gingivalis, Prevotella intermedia and Bacteroides fragilis. Oral Microbiol Immunol, 1996. 11(2): p. 79-87.

53. Moritz, A.J., et al., Immunization with Porphyromonas gingivalis cysteine protease: effects on experimental gingivitis and ligature-induced periodontitis in Macaca fascicularis. J Periodontol, 1998. 69(6): p. 686-97.

54. Nazarenko, I.A., S.K. Bhatnagar, and R.J. Hohman, A closed tube format for amplification and detection of DNA based on energy transfer. Nucleic Acids Res, 1997. 25(12): p. 2516-21.

55. Lindhe, J., A.D. Haffajee, and S.S. Socransky, Progression of periodontal disease in adult subjects in the absence of periodontal therapy. J Clin Periodontol, 1983. 10(4): p. 433-42.

56. Robins, J.M., An analytic method for randomized trials with informative censoring: Part II. Lifetime Data Anal, 1995. 1(4): p. 417-34.

57. Robins, J.M. and D.M. Finkelstein, Correcting for noncompliance and dependent censoring in an AIDS Clinical Trial with inverse probability of censoring weighted (IPCW) log-rank tests. Biometrics, 2000. 56(3): p. 779-88.

58. Little RJA, R.D., Statistical Analysis with Missing Data. 1987, New York: Wiley.

59. McGuigan KA, E.P., Hays RD, Bell RM, Adjusting for attrition in school-based samples: Bias, precision, and cost trade-offs of three methods. Evaluation Review, 1997. 21: p. 554-567.

60. Cohen, M.E., Analysis of ordinal dental data: evaluation of conflicting recommendations. J Dent Res, 2001. 80(1): p. 309-13.

61. Fleiss, J.L., et al., Statistical properties of some clinical measures of gingivitis and periodontitis. J Periodontol, 1990. 61(4): p. 201-5.

62. Scharfstein DO, R.A., Robins JM, Adjusting for nonignorable drop-out using semiparametric nonresponse models (with discussion). J. Amer. Stat. Assn, 1999. 94: p. 1096-1146.



I.2. Organization and Administration (Including Study Roster)
I.2.1. Participating Units
The Administrative Center, directed by Dr. Bryan Michalowicz, is housed within the Minnesota Oral Health Clinical Research Center at the University of Minnesota (UMN) School of Dentistry.
The Data Coordinating Center, directed by Dr. James Hodges, will be operated by the Coordinating Centers for Biometric Research in the Division of Biostatistics at the UMN.
The Plaque Analysis Laboratory, directed by Dr. Joel Rudney, is located at the UMN School of Dentistry.
The Serum Analysis Laboratory, directed by Dr. Jeffrey Ebersole, is located at the University of Kentucky School of Dentistry.
There are four clinical enrollment sites: Jackson Medical Mall, University of Mississippi, Jackson, MS (PI: Dr. William Buchanan); Hennepin County Medical Center, Minneapolis, MN (PI: Dr. Anthony DiAngelis); University of Kentucky, Lexington KY (PI: Dr. John Novack); and Columbia University / Harlem Hospital, New York, NY (PI: Dr. Panos Papapanou).
I.2.2. Study Organization and Administration
The study organization and administrative structure are shown in Figures 2a and 2b. The Administrative Center is located at the University of Minnesota. Dr. Michalowicz is the Study Chairperson and Chairman of Steering Committee. As such, he will have overall administrative, financial and scientific responsibility for the trial. He will report to the Data and Safety Monitoring Board and to NIH Staff, who will oversee monitoring activities as outlined in NIH/NIDCR Policies and Procedures.
I.2.2.1 The Steering Committee. The Steering Committee will advise the Chair on study related matters. It is the main leadership committee of the trial and is responsible for its overall direction. The Steering Committee will be responsible for:


  • the general design and conduct of the trial and preparation of essential documents including the protocol, manual of operations and data collection forms

  • reviewing and approving data collection procedures

  • approving changes in study procedures as appropriate

  • creating, making appointments to, and disbanding subcommittees

  • allocating resources based on competing study demands

  • reviewing study progress and implementing steps needed to allow the trial to meet its objectives

  • reviewing and implementing NIDCR-approved recommendations from the DSMB.

The Steering Committee consists of the Study Chair, the Principal Investigator (or Co-Principal Investigators) and Obstetric Investigator from each enrollment site, the Director of the Data Coordinating Center, Dr. Amos Deinard from the Department of Pediatrics at the UMN Medical School, and the NIDCR program official.

Current Steering Committee members are:


  • Dr. Bryan Michalowicz, Minnesota Oral Health Clinical Research Center, University of Minnesota (Chair)

  • Drs. William Buchanan and James Bofill at the University of Mississippi

  • Drs. John Novak and Jeff Ferguson, University of Kentucky

  • Drs. Anthony DiAngelis and Virginia Lupo, HCMC, Minneapolis, Minnesota

  • Drs. Panos Papapanou and Stephen Matseoane, Columbia/Harlem Hospital , New York, New York

  • Dr. James Hodges, University of Minnesota (Director, DCC)

  • Dr. Amos Deinard, University of Minnesota

  • Dr. Richard Mowery, National Institute of Dental and Craniofacial Research

The Committee will meet annually in Minneapolis, MN. Other meetings will be held by bimonthly teleconference call as needed. Email will be a primary means of daily communication between members of the Steering Committee.






I.2.2.2. Enrollment site personnel. The Enrollment Site Principal Investigator will have overall administrative, scientific and fiscal responsibility for that site. The Study Coordinator will be responsible for its day-to-day operation, coordinating all activities of the Principal Investigator and other study personnel and ensuring that data are collected in an accurate, timely and efficient manner. The on-site study coordinators will work closely with the Study Manager of the Data Coordinating Center to ensure that all data forms are completed accurately and in a timely manner.
The Enrollment Site PIs are:


  • Anthony J. DiAngelis, DMD, MPH, Chief of Dentistry, Hennepin County Medical Center, Minneapolis, MN




  • William Buchanan, D.D.S., M.Md.Sc., Department of Periodontics, University of Mississippi School of Dentistry, Jackson, MS




  • M. John Novak, BDS, LDS, MS, PhD, Center for Oral Health Research, University of Kentucky College of Dentistry, Lexington, KY




  • Panos N. Papapanou, DDS, PhD and Dennis A. Mitchell DMD, Division of Periodontics Columbia University School of Dental & Oral Surgery, New York, NY


I.2.2.3. Data Coordinating Center (DCC). James S. Hodges, PhD is the DCC's director. He is a voting member of the Steering Committee and developed the study's statistical aspects. He will supervise the DCC’s operations and present reports to the DSMB. He will supervise all statistical monitoring, will direct and actively participate in preparations for DSMB meetings and supervising preparation of other reports. He will direct statistical analyses of study data, developing new methods as needed, and co-author publications.
The DCC is not just a service bureau; it has taken and will take a leadership role in the study's design and scientific conduct. Communication, cooperation, and frequent interaction with investigators are essential ingredients in executing DCC responsibilities. Accordingly, the DCC's responsibilities involve most aspects of the study and include: participating in developing and modifying the study; forms design; setting up and maintaining data-collection procedures and documenting them in the Manual of Operations; training data collectors; setting up and operating the randomization system; executing data-collection and data-management procedures; producing and distributing reports, including DSMB reports and reports on enrollment, follow-up, protocol adherence, and data quality; analyzing study data for reports, publications, presentations and other needs; and assisting in writing publications and presentations.

I.2.3. Study Roster
University of Minnesota
Study Chair

Bryan Michalowicz, DDS, MS

Department of Preventive Sciences

School of Dentistry

17-116 Moos Tower

515 Delaware Street SE

Minneapolis, MN 55455

Phone: 612-625-6981

Fax: 612-626-2652

Email: micha002@umn.edu


Pediatric Investigator

Amos S. Deinard, MD, MPH

Ruben-Bentson Chair

Pediatric Community Health

Department of Pediatrics

MMC 85


420 Delaware St. SE

Minneapolis, MN 55455

Phone: 612-638-0700, ext. 212

Pager: 612-899-2147

Fax: 612-627-4205

Email: deina001@umn.edu


Microbiology Laboratory
Department of Oral Sciences

School of Dentistry

17-252 Moos Tower

515 Delaware St. SE

Minneapolis, MN 55455
Joel Rudney, PhD, Director

Phone: 612-624-7199

Fax: 612-626-2651

Email: jrudney@umn.edu



Data Coordinating Center:

1   2   3   4   5   6   7   8   9   ...   18


The database is protected by copyright ©dentisty.org 2016
send message

    Main page