Optimal combination of estimating equations in the analysis of multilevel nested correlated data.

Multilevel nested, correlated data often arise in biomedical research. Examples include teeth nested within quadrants in a mouth or students nested within classrooms in schools. In some settings, cluster sizes may be large relative to the number of independent clusters and the degree of correlation may vary across clusters. When cluster sizes are large, fitting marginal regression models using Generalized Estimating Equations with flexible correlation structures that reflect the nested structure may fail to converge and result in unstable covariance estimates. Also, the use of patterned, nested working correlation structures may not be efficient when correlation varies across clusters. This paper describes a flexible marginal regression modeling approach based on an optimal combination of estimating equations. Particular within-cluster and between-cluster data contrasts are used without specification of the working covariance structure and without estimation of covariance parameters. The method involves estimation of the covariance matrix only for the vector of component estimating equations (which is typically of small dimension) rather than the covariance matrix of the observations within a cluster (which may be of large dimension). In settings where the number of clusters is large relative to the cluster size, the method is stable and is highly efficient, while maintaining appropriate coverage levels. Performance of the method is investigated with simulation studies and an application to a periodontal study.

A Prospective Cohort Study of Bisphenol A Exposure from Dental Treatment.

Laboratory studies show that bisphenol A (BPA) leaches from bisphenol A-glycidyl methacrylate (bisGMA)-based dental materials. We aimed to quantify the extent to which children are exposed to BPA from dental treatment with bisGMA materials, by amount of treatment and type of sedation. We hypothesized that posttreatment urinary BPA (uBPA) concentrations would be higher among patients with more surfaces treated with bisGMA-based materials and among patients receiving general anesthesia compared with pretreatment concentrations. We conducted a prospective cohort study in 211 children, 4 to 12 y old, who had no prior resin-based dental treatment. We measured uBPA concentrations twice before treatment and at 2 d and 1, 4, and 16 wk posttreatment. We abstracted treatment data (surfaces treated) from the chart. We generated descriptive statistics and compared pre- and posttreatment uBPA concentrations using generalized estimating equations. Participants were 51% female, 46% non-White, and 74% publicly insured. The median age was 6 y. The mean number of tooth surfaces exposed to BisGMA materials (composites/sealants) was 7.5 (SD 5.3). Overall, uBPA concentrations were 86% higher (95% confidence interval [CI] 42% to 143%, P < 0.001) at 2 d posttreatment compared with pretreatment concentrations. The uBPA concentrations 2 d posttreatment versus pretreatment tended to be higher (112%, 95% CI 53% to 194%) among those receiving treatment on >4 surfaces than those receiving treatment on ≤4 surfaces (50%, 95% CI -2% to 130%). Two days after treatment, uBPA was significantly higher than pretreatment concentrations in children receiving nitrous oxide but not in those receiving general anesthesia. Among all findings, uBPA concentrations returned to baseline by 4 wk. Children experience short-term increases in BPA from dental treatment. The impact of relatively high, short-term BPA exposure on child health is unknown. Given the widespread use of BisGMA-based dental materials and that chronic low-dose BPA exposure may adversely affect child health, strategies that minimize BPA exposure could potentially improve child health.

Dental amalgam and antibiotic- and/or mercury-resistant bacteria.

Mercury emitted from dental amalgam may select for increased numbers of antibiotic- or mercury-resistant commensal bacteria in patients and increase their risk for bacterial diseases that are resistant to common therapies. We hypothesized that the presence of dental amalgams would increase the level of mercury-, tetracycline-, ampicillin-, erythromycin-, or chloramphenicol-resistant oral and urinary bacteria as compared with levels in children receiving composite fillings. Samples were collected at baseline, 3-6 months after the initial dental treatment, and annually for 7 years of follow-up. There were no statistically significant differences between treatment groups in the numbers of bacteria growing on antibiotic- or mercury-supplemented plates. This study provided no evidence that amalgam fillings on posterior teeth influenced the level of antibiotic- or mercury-resistant oral or urinary bacteria as detected by culture.

Site-Specific Neutrophil Migration and CXCL2 Expression in Periodontal Tissue.

The oral microbial community is the best-characterized bacterial ecosystem in the human host. It has been shown in the mouse that oral commensal bacteria significantly contribute to clinically healthy periodontal homeostasis by influencing the number of neutrophils that migrate from the vasculature to the junctional epithelium. Furthermore, in clinically healthy tissue, the neutrophil response to oral commensal bacteria is associated with the select expression of the neutrophil chemokine CXCL2 but not CXCL1. This preliminary study examined the contribution of commensal bacteria on neutrophil location across the tooth/gingival interface. Tissue sections from the root associated mesial (anterior) of the second molar to the root associated distal (posterior) of the second molar were examined for neutrophils and the expression of the neutrophil chemokine ligands CXCL1 and CXCL2. It was found that both the number of neutrophils as well as the expression of CXCL2 but not CXCL1 was significantly increased in tissue sections close to the interdental region, consistent with the notion of select tissue expression patterns for neutrophil chemokine expression and subsequent neutrophil location. Furthermore, mice gavaged with either oral Streptococcus or Lactobacillus sp. bacteria induced a location pattern of neutrophils and CXCL2 expression similar to the normal oral flora. These data indicate for the first time select neutrophil location and chemokine expression patterns associated with clinically healthy tissue. The results reveal an increased inflammatory load upon approaching the interproximal region, which is consistent with the observation that the interproximal region often reveals early clinical signs of periodontal disease.

Circadian variation in urinary excretion of bone collagen cross-links.

Bone resorption can be evaluated by measuring the urinary excretion of collagen type I cross-linked N telopeptides (NTx). Since it is difficult to obtain (and verify) 24 h urine collections from patients, untimed spot urines are more practical. Such measurements, however, need correction for urine dilution and potentially may vary with collection time since a circadian rhythm in bone metabolism has been reported. This study examined cross-link excretion in urine voids serially collected during a 24 h period from subjects living their normal daily routine (as opposed to a controlled hospital setting). This mimics the situation for walk-in patients visiting a clinician and providing a spot urine. A total of 35 dentists (20 males, 15 females) collected all urine voids separately over a 24 h period. Urines were analyzed for creatinine and NTx. The effects of time of day on the excretion rates of these metabolites (in nmol/h) and on the cross-link:creatinine ratio were assessed. A circadian rhythm was evident in the excretion rate of creatinine with a peak in the late afternoon (18% higher than the 24 h mean, p = 0.0004). The NTx excretion rate peaked in the morning (9% higher than the 24 h mean) but this latter rhythm was not statistically significant (p = 0.31). The NTx:creatinine ratio fell during the day from a high (122% of the 24 h mean) in the early morning to a low in the early evening. This rhythm in the NTx:creatinine ratio in untimed spot urines was statistically significant (p < 0.0001). In conclusion, the NTx:creatinine ratio in spot urines from adult outpatient subjects showed a significant circadian rhythm. Variations in creatinine excretion were the primary cause. Time of day should, therefore, be taken into account when comparing test results of spot urines with normal ranges or with other samples from the same subject.

AMESFIT: a microcomputer program for fitting linear-exponential dose-response models in the Ames Salmonella assay.

The Ames Salmonella/microsome assay remains the most widely used microbial test for genotoxicity. In this article, we describe a microcomputer program developed to fit a linear-exponential dose-response model to Ames assay data for established mutagens. The model includes a linear term to describe the mutagenic effects of the test agent at low to moderate doses and an exponential attenuation factor to accommodate downturns at high doses due to cytotoxicity. Quasi-likelihood methods are used to obtain estimates of the unknown model parameters, thereby avoiding the need to fully specify the distribution of the experimental data. This method of estimation also allows for extra-Poisson variation that is characteristic of counts of mutant colonies of bacteria observed in the Ames assay. The particular linear-exponential model used here was developed for use in the analysis of a recent large-scale collaborative trial using the Ames assay sponsored by the International Programme on Chemical Safety. The use of our program is illustrated using sample data sets taken from that collaborative study.

Between-subject and within-subject statistical information in dental research.

The evaluation of risk factors in dental research frequently uses observations at multiple sites in the same patient. For this reason, statistical methods that accommodate correlated data are generally used to assess the significance of the risk factors (e.g., generalized estimating equations, generalized linear mixed models). In applications of these methods, it is typically assumed (implicitly, if not explicitly) that between-subject and within-subject comparisons will produce the same estimated effect of the risk factor. When between- and within-subject comparisons conflict, the statistical methods can give biased estimates or results that are difficult to interpret. For illustration, we present two examples from periodontal disease studies in which different statistical methods give different estimates and significance levels for a risk factor. Statistical analyses in dental research should assess whether different sources of information give similar conclusions about risk factors or treatments.

Evaluating the validity of probing attachment loss as a surrogate for tooth mortality in a clinical trial on the elderly.

Most periodontal trials are based on the assumption that the superior treatment, as judged by short-term intangible changes in probing attachment levels (the surrogate), is also the treatment most likely to affect tooth mortality. This assumption is valid if: (1) the surrogate is informative about tooth mortality, and (2) the surrogate captures a substantial proportion of the treatment effect on tooth mortality (e.g., > 50% or 75%). The goal of this study was to evaluate whether both conditions were satisfied in a randomized controlled trial (RCT) of elders at high risk for dental diseases. The results suggested that the first condition for a valid surrogate was satisfied: Both one- and two-year changes in probing attachment level were informative about tooth mortality risk. A 1-mm loss measured over a one-year period was associated with a 56% increased tooth mortality risk (relative risk = 1.56; 95% confidence interval, 1.08 to 2.26; p = 0.017); a 1-mm loss measured over a two-year period was associated with a 102% increased risk for tooth mortality (relative risk = 2.02; 95% confidence interval, 1.26 to 3.25; p = 0.004). The second condition necessary for a valid surrogate could not be confirmed in the present trial. With 95% confidence, it was concluded that one-year changes in probing attachment level measurements did not capture a significant proportion of the treatment effect (point estimate, 6%; 95% confidence interval;-38% to 53%). No useful statements could be made regarding the proportion of treatment effect captured by two-year changes, due to the width of the confidence interval (point estimate, 18%; 95% confidence interval;-151% to 140%). It is concluded that (1) the evidence surrounding the one-year change in probing attachment level indicates that it can be ruled out as being anything more than a weak surrogate marker for tooth mortality, and (2) further research is required to study the validity of two-year change in probing attachment level as a surrogate marker. Due to characteristics of the population and the treatments investigated, the generalizability of these findings to other RCTs is questionable.

The estimation of caries prevalence in small areas.

National surveys have been effective for the estimation of caries prevalence in broad regions of the US. However, it is unclear if data from such surveys can be used to estimate prevalences in small areas such as states or counties because of small sample sizes within individual areas. In this study, we applied specialized statistical methods to the estimation of small-area caries measures using data from an oral health survey conducted in the State of Washington. Dental exams to assess caries and the presence of sealants and fluorosis were performed on 2921 third-grade students in 84 public schools selected by a stratified random sample from all 39 counties in the state. Statistical methods for small-area estimation were used to estimate disease and sealant utilization measures for each of the counties. Adjustment was made for covariates measured at the school level, including ethnicity and the proportion of children in the Federally sponsored school lunch program. Substantial variability in disease and sealant utilization between counties was found. The estimated number of decayed and filled surfaces per child was 4.7 (inter-county range, 2.4 to 7.4). The estimated number of surfaces of untreated decay was 1.2 per child overall (range, 0.5 to 3.1). Thirty percent of the children had restorative treatment needs (range, 15 to 54%). The prevalence of sealants on one or more permanent molars was estimated to be 34% (range, 19 to 46%). Overall, only 8% of children showed evidence of fluorosis. The results demonstrate the usefulness of small-area estimation methods for oral health surveys.

Forty-five-year tooth survival probabilities among men in Oslo, Norway.

Changes in tooth survival probabilities over a person's lifetime have remained largely unexplored. The goal of this study was to evaluate changes in the 45-year tooth survival probabilities in a cohort of 565 Norwegian males who were examined in 1969 as young adults, and followed up into mid-life (examination years and sample sizes (n): 1971 (n = 381), 1973 (n = 292), 1975 (n = 245), 1981 (n = 228), 1988 (n = 202), and 1995 (n = 223). The results indicated that the tooth survival probabilities varied considerably both (i) among teeth within individuals, and (ii) over time. The 45-year survival probabilities for the 28 teeth fell into the following ranges: larger than 95% for incisors and cuspids; between 84% and 92% for premolars; and between 59% and 96% for molars. Over the first 4 post-eruptive decades, the tooth mortality risks (excluding orthodontic extractions) were: 1st decade, 2.0% (from 1.7 to 2.4%); 2nd decade, 0.2% (from 0.1 to 0.4%); 3rd decade, 0.6% (from 0.4 to 0.8%); and 4th decade, 1.1% (from 0.8 to 1.5%). The tooth mortality risks in the 2nd, 3rd, and 4th decades were probably somewhat underestimated (due to dropout bias), suggesting that the true underlying tooth mortality hazard function may have been V-shaped. The conclusions were that the tooth mortality hazard during the first 4 post-eruptive decades was bathtub-shaped and that it varied considerably among teeth within individuals.

Within-subject coronal caries distribution patterns: an evaluation of randomness with respect to the midline.

The distribution of caries among homologous surfaces can exhibit three possible patterns: random, aggregated, or regular. In a random caries pattern, caries lesions are randomly distributed among homologous surfaces. An aggregated caries pattern is distinguished by the aggregation of lesions on one side of the mouth or the other to a greater extent than would be expected by chance alone. For a regular caries pattern, the left-right distribution of lesions is more homogenous than would be expected by chance alone. A test statistic based on the left-right distribution of caries lesions among discordant homologous pairs was developed to investigate which of these three caries patterns is present in a representative sample of the adult United States population. The data originated from the National Survey of Oral Health in the US (Employed Adults), 1985-1986. Of the 15,132 subjects studied, 12,776 subjects had 2 or more decayed or filled teeth. (At least 2 carious or filled teeth are required for detection of patterns.) Approximately 50% of these subjects (n = 6,439) had two or more discordant homologous tooth pairs. With these tooth pairs, the hypothesis of a random caries pattern was rejected in favor of an aggregated caries pattern (p < 0.0001). Similar findings were obtained with discordant homologous surface pairs. This aggregation of caries on one side of the mouth or the other may be due to genetic, infectious, and/or environmental factors.

Sources of variation in the mutagenic potency of complex chemical mixtures based on the Salmonella/microsome assay.

Twenty laboratories worldwide participated in a collaborative trial sponsored by the International Programme on Chemical Safety on the mutagenicity of complex mixtures as expressed in the Salmonella/microsome assay. The U.S. National Institute of Standards and Technology provided homogeneous reference samples of urban air and diesel particles and a coal tar solution to each participating laboratory, along with samples of benzo[a]pyrene and 1-nitropyrene which served as positive controls. Mutagenic potency was characterized by the slope of the initial linear component of the dose-response curve. Analysis of variance revealed significant interlaboratory variation in mutagenic potency, which accounted for 57-96% of the total variance on a logarithmic scale, depending on the sample, strain and activation conditions. Variation among replicate extractions of organic material (required for the air and diesel particles) and among replicate bioassays within the same laboratory was also appreciable. The average potencies for air and diesel particles in laboratories using Soxhlet extracts were not significantly different from those in laboratories using sonication, although there was larger interlaboratory variation for the Soxhlet method. Repeatability (which approximates the coefficient of variation within laboratories) ranged from 18 to 40% for air and diesel particles extracted using sonication, depending on the strain and activation conditions. Repeatability of Soxhlet-extracted air and diesel particles, however, ranged from about 37 to 89% including outliers and from about 11 to 31% excluding outliers. Repeatability of the coal tar sample and the 2 positive controls was in the range 18-34%. Reproducibility (which approximates the coefficient of variation between laboratories) was generally at least twice repeatability, and exceeded 100% for Soxhlet-extracted air and diesel particles, as well as 1-nitropyrene. Reanalysis of the data omitting observations of more than 1500 revertants/plate generally had little effect on these results. Elimination of outlying observations had limited impact, with the exception of Soxhlet-extracted air and diesel particles. In this case, reproducibility of bioassay results was notably improved, due largely to the omission of results for replicate extractions which varied more than 5-fold within one laboratory. Normalization of the log potency slopes for the mixtures by the corresponding slopes for benzo[a]pyrene tended to reduce this variation, although variation was increased after normalization by 1-nitropyrene. Adjustment for the percentage of organic matter extracted from the air and diesel particulate samples had little effect on variation for sonication-extracted particles, whereas variation was reduced for diesel particles and increased for air particles for Soxhlet.

Reliability of individual differences in initial sensitivity and acute tolerance to nitrous oxide hypothermia.

On average, the hypothermia exhibited by rats receiving 60% nitrous oxide (N2O) eventually abates despite the continued inhalation of the drug (i.e., acute tolerance develops). However, large individual differences occur in both the magnitude of hypothermia achieved and the degree of acute tolerance that develops. To determine whether the degree of temperature loss and subsequent recovery during N2O administration are reliable characteristics of an individual, we measured intraperitoneal temperature via telemetry in 77 Long-Evans rats that each received 60% N2O for 5 h during two sessions separated by 14 days. Good intersession reliability (Pearson's r) was observed for simple change and adjusted change scores for both initial N2O temperature sensitivity (.61 < or = r < or = .62), and acute tolerance development (.46 < or = r < or = .52). In a separate experiment, three groups of rats were selected based on their individual body temperature patterns during an initial N2O administration: (1) insensitive to N2O hypothermia (n = 8); (2) marked hypothermia followed by acute tolerance development (n = 6); and (3) marked hypothermia followed by little acute tolerance development (n = 6). When retested 10 days later, each group exhibited a body temperature profile similar to that observed during the initial N2O exposure. Thus, the temperature profile observed during a rat's initial exposure to 60% N2O reflects a reproducible response for that animal.

Nitrous oxide-induced hypothermia in the rat: acute and chronic tolerance.

Although inhalation of nitrous oxide (N2O) causes hypothermia in rats, there is a paucity of information as to whether tolerance develops to this effect. The purpose of this study was to determine whether tolerance to N2O hypothermia develops within a single administration as well as over repeated administrations. Temperature was measured telemetrically by implanting intraperitoneal thermal sensors/transmitters in male Long-Evans rats. Experimental rats received an initial 2-h exposure to 60% N2O and became hypothermic relative to controls breathing placebo gas. Only a few rats demonstrated evidence of acute tolerance over the 120 min. Over the next 10 days, the experimental rats received five additional 30-min exposures to 60% N2O and five 30-min exposures to placebo while the control rats received only placebo gas exposures. Chronic tolerance developed to N2O hypothermia over these repeated administrations. A test for Pavlovian drug conditioning found no evidence that conditioned temperature effects contributed to chronic tolerance development. In a second experiment, naive rats were given a 380-min exposure to 60% N2O and a 380-min exposure to placebo gas in a counterbalanced order. Acute tolerance did develop to N2O hypothermia, with the recovery of temperature beginning after a mean of 141 min of gas administration. Hence, both acute and chronic tolerance develop to N2O's hypothermic effects in rats.

Validity of spot urine samples as a surrogate measure of 24-hour porphyrin excretion rates. Evaluation of diurnal variations in porphyrin, mercury, and creatinine concentrations among subjects with very low occupational mercury exposure.

Porphyrin, mercury, and creatinine levels in single-void urine specimens ("spot samples") were compared with calculated 24-hour urine concentrations among 146 (77 male and 69 female) practicing dentists who are participating in an ongoing study of urinary porphyrin changes as a biomarker of mercury body burden. All subjects had urinary mercury concentrations < or = 5 micrograms/L, a level comparable to that of the general US population and which is below that determined to be associated with mercury-induced changes in urinary porphyrin excretion rates. The results confirmed previous findings of no significant diurnal variation in any of the porphyrin levels normally found in urine or in total porphyrin levels combined among males but describe significant diurnal variations in most porphyrin levels as well as concentrations of total porphyrins combined among females. Similarly, no evidence of large diurnal variation in mercury excretion among males was apparent, whereas significant diurnal variation in the mercury excretion rate among females was found. Creatinine adjustment of porphyrin or mercury concentrations had no significant effect on these findings. Moreover, no evidence of diurnal variation in urinary creatinine excretion among either male or female subjects was obtained, despite substantial between-subject variability in this parameter. These results support the view that spot urine samples may be utilized to derive reasonably accurate estimates of 24-hour porphyrin and mercury excretion rates in male subjects. In contrast, time of day appears to be of considerably greater importance when spot samples are utilized as 24-hour estimates of either porphyrin or mercury excretion rates among females. Additionally, time of day may be an important consideration in studies involving serial (repeated) porphyrin or mercury measurements using spot urine samples, irrespective of gender distribution of study subjects.

Evaluating the burst hypothesis at a site-specific level using the lack-of-fit test.

It has been hypothesized that periodontal disease progresses by means of sudden losses of periodontal attachment surface area. Obtaining reliable tests of this burst hypothesis has proven to be difficult; the signal (true model of disease progression) often gets lost in the noise. The purpose of this study was to determine how reliably we could distinguish sudden changes from linear disease progression at a site using a time series of clinical attachment levels. Specifically, the following question was investigated: If, in reality, disease progresses by means of sudden changes in clinical attachment level (bursts), and a linear model is fitted to these data, what is the likelihood of rejecting the linear model using the lack-of-fit test? This likelihood was determined as a function of the probing measurement error (range: 0.2 to 1.0 mm) and the number of clinical examinations over time. The results suggested that bursts of 2 mm or smaller cannot be reliably distinguished from linear disease progression using the lack-of-fit test, except under unusual clinical circumstances. Under typical clinical circumstances, burst sizes needed to be 3 to 5 mm in order to be reliably distinguished from linear disease progression. These results are probably overly optimistic. The ability to verify the burst hypothesis at the site level is likely to be even less than our results indicate because of various assumptions that were required. We conclude that the lack-of-fit test will reliably reject the linear model at a site-specific level only if true disease progresses in such a fashion that a handful of sudden changes leads to a tooth mortality event.

Non-surgical periodontal therapy and tooth loss. A cohort study.

BACKGROUND: No reliable evidence is available regarding the effect of periodontal therapy on major disease endpoints such as tooth loss, edentulism, or quality of life. The primary objective of this study was to assess the association between tooth loss and the non-surgical periodontal treatment history of 1,021 members of the Kaiser Permanente Dental Care Program. METHODS: Tooth loss rates were estimated using Poisson regression models, adjusting for some of the potentially confounding variables such as initial disease severity and extent. RESULTS: Continuous non-surgical therapy (one or more non-surgical procedures performed during 3 successive years), as opposed to no therapy during such a 3-year period, reduced the subsequent tooth mortality rate by 58% (relative rate, 0.42; 95% confidence interval, 0.29-0.61). Intermittent non-surgical therapy reduced the tooth mortality rate by 48% (RR = 0.52; 95% confidence interval, 0.34-0.80). As the number of non-surgical procedures increased, tooth loss rates decreased. CONCLUSIONS: These findings suggest that non-surgical periodontal therapy may be associated with a substantial reduction in tooth mortality. Different study designs and populations are needed to confirm these findings.

The informativeness of attachment loss on tooth mortality.

BACKGROUND: In periodontal clinical trials, clinical attachment level measurements are commonly used as surrogates for tooth loss. Conclusions regarding treatment efficacy in these trials are valid if: 1) the surrogate is informative on tooth loss, and 2) the surrogate captures the effect of treatment on tooth loss. The goal of this study was to evaluate the first criterion: Are serial clinical attachment loss measurements informative on overall tooth mortality? METHODS: Young Norwegian men (aged 17 to 35) were first examined in 1969 (n=565) and followed for 26 years with examinations in 1971 (n=381), 1973 (n=292), 1975 (n=245), 1981 (n=228), 1988 (n=202), and 1995 (n=223). Several aspects of the serial attachment loss measurements were related to tooth mortality risk using statistical models that take into account the time-dependent changes of the clinical attachment loss measurements. RESULTS: The results provided evidence that moderate attachment losses were informative on tooth mortality. Both the lifetime cumulative attachment loss, as well as attachment loss since young adulthood, of > or = 2 mm or > or = 3 mm was informative on tooth mortality. Tooth mortality risk increased as the attachment loss increased; loss > or = 3 mm at the buccal or mesial site increased tooth mortality risk, by 91% (relative risk, 1.91; 95% confidence interval, 1.01-3.60) and 270% (RR, 3.70; 95% CI, 1.83-7.49), respectively. CONCLUSIONS: We concluded that clinical attachment loss was moderately informative on overall tooth mortality in this Norwegian population. Since this finding has now been demonstrated in 3 different populations, the focus of further research should be on evaluating whether the second criterion for a valid surrogate is satisfied: Does clinical attachment loss capture the effect of periodontal treatments on tooth loss?

Factors associated with caries incidence in an elderly population.

The purpose of this paper was to identify baseline factors associated with future caries development in older adults (age 60+) during a 3-year study period. Poisson regression analysis was used to determine the association between potential risk factors and disease incidence. The significant factors associated with high coronal caries incidence rates were high baseline root DMFS (P<0.001), high counts of mutans streptococci and lactobacilli (P=0.036), male gender (P=0.007), and Asian ethnicity (P=0.002). These factors had small to moderate effects on incidence rates, with relative risk values of approximately 1.2 to 2. The significant factors associated with higher disease incidence on root surfaces were baseline coronal DMFS (marginally significant, P=0.078), high bacterial counts (P=0.002), and Asian ethnicity (P=0.009). The predictive value of the models was low for both coronal and root caries. This result may be because this population had a higher than usual caries incidence rate, making discrimination among these caries-active individuals difficult. The current study affirmed the value of baseline DMFS and salivary variables to modeling caries incidence and introduced ethnicity as a variable useful for the study of dental caries in older adults.

Effects of Le Fort I osteotomy on human gingival and pulpal circulation.

The maxillary blood flow during the first 24 h following Le Fort I osteotomy was evaluated by laser Doppler flowmetry. Pulpal blood flow was recorded from two maxillary incisors and gingival blood flow was assessed from a site slightly apical to the interdental papilla of the maxillary central incisors of 12 patients receiving Le Fort I osteotomy, nine control patients receiving mandibular osteotomy, and 10 nonsurgical control subjects without orthodontic appliances. Measurements were made before surgery and at time intervals between 0-8, 8-16, and 16-24 h after surgery. The nonsurgical controls were tested at similar intervals. Gingival and pulpal blood-flow measurements did not change over time in the nonsurgical control group. Presurgical blood-flow values did not differ between the two surgical groups. Following surgery, mean gingival (but not pulpal) blood flow was significantly lower for patients treated with Le Fort I osteotomy than for patients treated with mandibular osteotomy. Follow-up examinations revealed that one patient receiving Le Fort I osteotomy experienced loss of gingiva and bone around both central incisors. This patient had one of the largest reductions in gingival blood flow.