Dissertation committee:
Λουτράδης Δημήτριος, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Δρακάκης Πέτρος, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ, Επιβλέπων
Στεφανίδης Κωνσταντίνος, Αναπληρωτής Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Παππά Καλλιόπη, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Ντόμαλη Αικατερίνη, Επίκουρη Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Γρηγοριάδης Θεμιστοκλής, Επίκουρος Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Θωμάκος Νικόλαος, Επίκουρος Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Summary:
Premature birth, low birth weight and preeclampsia are among the most important pregnancy complications.The objective of the present study is to determine the relationship between mother’s periodontal disease and incidence of preeclampsia, premature birth and low birth weight. For the purposes of this study 359 pregnant women were examined while they visited Alexandra hospital for their antenatal care or while being hospitalized for various reasons. The participants after being informed of the purposes of this study and if they fulfilled the selection criteria were asked for their written consent and a means of communication (telephone number), in order to give additional information if needed about later complications or pregnancy outcomes if labor would take place in other hospital than Alexandra hospital. Selection criteria were women to be pregnant, if underage to be given the guardian’s consent for participation, to have at least 16 teeth and to be of Caucasian heritage. Women were excluded if they required prophylactic antibiotics before dental treatment and if they had a current multiple pregnancy. For the analysis of the relationship between periodontal disease and preeclampsia in the exclusion criteria history of preeclampsia and hereditary thrombophilia were added. The study was conducted in two phases. In the initial phase the participants were selected, a questionnaire was filled in and a thorough clinical periodontal exam was completed. In the second phase after the labor participants were contacted by the means of communication they had provided (telephone number) and were asked the date of delivery, the duration of gestation, if labor took place in Alexandra hospital or in another hospital, the newborn’s weight at delivery, if there had been any pregnancy complications after the exam that took place in the first phase and if labor took place by caesarian section or vaginal delivery. The clinical examination included a full-mouth periodontal examination (determining probing depth and attachment loss at six sites on all of the teeth) and bleeding on probing index. After the clinical
examination aspects of dental history, which indicated local inflammation and could potentially raise the systemic inflammatory burden, were registered. Alexandra hospital’s medical records verified the data that were collected. Also from the hospital’s medical records data about gestation duration, birth weight, pregnancy pathology and type of delivery were verified.
Finally in this study complete data from 300 pregnant women were gathered, 14-45 years old, with mean age 30,7±6,3 years old. The mean gestational age at which the clinical examination took place was 31,3±8,3 weeks, calculated based on last menstrual period. Mean pregnancy duration was 37,9±2,8 weeks. Mean birth weight was 2942,2±712,3 grams. Periodontitis prevalence was depending on periodontitis definition 39-94%. Regarding educational level 10,67% had incomplete primary education, 9% had primary education, 25,33% had secondary education, 14,33% had post high school education and 40,67% had higher education.18,33% were current smokers, 18,67% had quitted smoking less than a year before they were pregnant and the rest 63% were regarded as no smokers.
For 50,67% this was their first pregnancy, 9,6% had a history of preterm birth, 23% had history of miscarriage and 9,3% had history of low birth weight.16% had been diagnosed with diabetes mellitus. 29% had been diagnosed with urinary tract infection during pregnancy and 36% had been prescribed with antibiotic treatment during pregnancy.48,33% of the participants reported dental history of inflammation of dental aetiology, 19,67% reported inflammation of dental aetiology during pregnancy, 18,33% had teeth which should be endodontically treated or extracted (pulp chamber exposure) and 19% had semi impacted teeth.
Regarding the relationship of maternal periodontal disease with the incidence of preeclampsia the results of this study were the following:
Univariate analysis
• pregnant women with at least 5 sites with probing depth of 5 mm or more (periodontitis definition 6) were 256% more likely to develop preeclampsia (chi-square test p=0,010 and p=0,026 with Yates correction OR:3,56, 95% CI:1,29-9,84 // kendall’s tau-b test p=0,083).
• there was a marginal statistical significance in the relationship of periodontitis definition 2 with preeclampsia (p=0,051)
• there was a statistical significance for the relationship of periodontitis definition 2 with preeclampsia for the age group 31-35 (p=0,017)
• there was a statistical significance for the relationship of periodontitis definition 5 with preeclampsia for the age group 31-35 (p=0,042)
• there was a tendency towards statistical significance for the relationship of mean probing depth with preeclampsia, when test was performed for the means (p=0,0573) and the medians (p=0,072) of mean probing depth
Multivariate analysis
• binary logistic regression analysis for the relationship of preeclampsia with the potential risk factors (for periodontitis definition 5 was used) resulted in the following models:
log(preeclampsia probability/1- preeclampsia probability)= -22,062 + 0,393 ΒΜΙ – 3,049 PRETERM BIRTH HISTORY – 0,006 BIRTH WEIGHT+ 0,681 WEEKS OF GESTATION (backward) log(preeclampsia probability/1- preeclampsia probability)= -22,062 + 0,363 ΒΜΙ – 3,049 PRETERM BIRTH HISTORY – 0,006 BIRTH WEIGHT + 0,681 WEEKS OF GESTATION (forward)
• the probability of preeclampsia incidence is increasing with the increase of mother’s BMI and with the increase of gestation duration. The probability of preeclampsia incidence is decreasing with history of preterm birth and with increase of birth weight
• the presence of at least 4 pockets did not alter the probability of preeclampsia incidence
• principal component analysis-PCA did not result in any relationship of periodontal parameters such as the number of sites with a pocket, the mean pocket depth and the number of sights with bleeding on probing and inflammation and bacteremia markers such as the presence of semi impacted teeth, the presence of destroyed teeth with pulp chamber exposure, the incidence of inflammation of dental aetiology during pregnancy and the history of tooth extraction with preeclampsia incidence
Critical review of the literature
• in the literature it is mentioned that periodontitis and preeclampsia share as common risk factors some genetic polymorphisms such as IL6-174G, IL10-592C, IL10-819C and TNFα-308A
• in the literature there is greater unanimity in the results regarding the relationship between periodontitis and preeclampsia than the results regarding periodontitis and preterm birth
• in the literature probing depth is associated with the incidence of preeclampsia independently of the presence of attachment loss (that is not the case for the relationship of probing depth with premature birth)
• in the literature periodontitis definitions which reached statistical significance regarding the relationship between periodontitis and preeclampsia used lower number of teeth for periodontitis definition than periodontitis definitions which reached statistical significance regarding the relationship between periodontitis and premature birth
• the observed differences in the literature about the relationship of periodontitis and preeclampsia and periodontitis and premature birth could be a result of the possible need for the bacteria to reach the maternal side of placenta in order to cause preeclampsia and the fetal side of the placenta in order to cause premature birth. The for mentioned difference in the literature could be an indirect confirmation of the theory of the hematogenous bacteria spread because travelling through blood vessels because bacteria travelling through blood vessels reach first the maternal side of placenta and then reach the fetal side of placenta. The probability of a greater dependence of the relationship of premature birth with genetic factors (because of the stronger relationship of premature birth with attachment loss) could exist due to the need for the bacteria, cytokines or bacteria toxins to cross the placenta barrier
• it is possible that some of preeclampsia incidents could be attributed in the combined effect of dysbiotic placenta microbiome, which is the result of hematogenous migration of bacteria from the oral cavity, and the immune response in pregnant women who have geneticpredisposition
Regarding the relationship of maternal periodontal disease with the incidence of preterm birth the results of this study were the following:
Univariate analysis
• pregnant women over 35 years old have statistically significant shorter gestation duration
• smoking results in statistically significant shorter gestation duration
• diabetes mellitus results in statistically significant shorter gestation
duration
• periodontitis neither affected gestation duration nor altered the number of newborn that
were born preterm
• periodontal parameters neither affected gestation duration (number of
sites with bleeding on probing, number of pockets with bleeding on probing) nor altered the number of newborn that were born preterm (mean probing depth, number of sights with bleeding on probing)
• inflammation of dental etiology during pregnancy did not alter the risk for preterm birth
• teeth with exposed pulp chamber which need extraction or endodontic therapy did not alter the risk for preterm birth
Multivariate analysis
• Linear regression analysis for the relationship of preterm birth with the potential risk factors (for periodontitis definition 5 was used) resulted in the following models:
WEEKS OF GESTATION= 38,029 – 1,927 HISTORY OF PRETERM BIRTH (stepwise)
WEEKS OF GESTATION= 39,352 – 0,055 AGE – 1,828 HISTORY OF PRETERM BIRTH + 0,628 PERIODONTITIS DEFINITION 5 (backward)WEEKS OF GESTATION= = 39,352 – 1,927 HISTORY OF PRETERM BIRTH (forward)
• the results from the three models of linear regression analysis agree
that history of preterm birth reduces gestation duration (approximately 2 weeks). The model, which resulted from backward linear regression analysis, proposed that the presence of at least 4 pockets in the oral cavity of a pregnant woman increases gestation duration 6/10 of a week which is of no clinical significance. Also the model, which resulted from backward linear regression analysis, proposed that one year increase in mother’s age, decreases gestation duration 0,055weeks.
• the presence of 4 pockets in mother’s periodontium did not alter the risk of preterm birth
• principal component analysis-PCA did not result in any relationship of periodontal parameters such as the number of sites with a pocket, the mean pocket depth and the number of sights with bleeding on probing and inflammation and bacteremia markers such as the presence of semi impacted teeth, the presence of destroyed teeth with pulp chamber exposure, the incidence of inflammation of dental etiology during pregnancy and the history of tooth extraction with preeclampsia incidence
Critical review of the literature
• in the literature attachment loss is associated more consistently with preterm birth than probing depth
• in the literature there is some evidence that the relationship between periodontitis and preterm birth is affected by genetic factors
• periodontitis and preterm birth share same genetic polymorphisms as common risk factors• in the literature there is sufficient number of studies which link periodopathic bacteria and dysbiotic placenta microbiome with preterm birth
• it is possible a part of premature births to occur due to the combined effect of dysbiotic placenta microbiome, which is a result of the hematogenous spread of bacteria from the oral cavity, and the immune response in pregnant women who are genetically predisposed, independently of if there is an active periodontal disease
• there is a need for more studies in order to clarify the possible relationship between maternal periodontitis and preterm birth
Regarding the relationship of maternal periodontal disease with the incidence of low birth weight the results of this study were the following:
Univariate analysis
• there is a statistically important difference between the incidence of smoking and birth weight
• there is a tendency towards statistical importance for the relationship of mean probing depth with low birth weight (median test, p=0,065). The median of mean probing depth for the group of normal weight infants was greater than the median of mean probing depth for the group of low birth weight infants
• maternal periodontitis ( regardless the definition that was used) neither altered the infant weight nor altered the number of low birth weight infants
• the periodontal parameters that were used neither altered the infant weight nor altered the number of low birth weight infants
• inflammation of dental etiology during gestation did not alter the risk of low birth weight infant
• teeth with exposed pulp chamber which need extraction or endodontic therapy did not alter the risk for low birth weight
Multivariate analysis
• binary logistic regression –forward method for the relationship ofpreterm birth with the potential risk factors (for periodontitis definition 5 was used) resulted in the following model:
log(low birth weight possibility/1 – low birth weight possibility) = 47,550 – 1,332 WEEKS OF GESTATION+ 0,558 SMOKING (YES)+ 1,727 SMOKING (YES BEFORE)
• binary logistic regression –forward method resulted that most important factor for incidence of low birth weight is preterm birth. If we solve the for mentioned equation we come to the conclusion that the weeks of gestation in which the possibility of low birth weight infant is equal to the possibility of infant of normal weight are 35,7 weeks for non smokers, 36,1 weeks for smokers, and 37 weeks for the women who quitted smoking less than a year before pregnancy. While for smokers and no smokers the probability of low birth weight infant and infant of normal weight infant are equal at approximately 36 weeks (4/10 week difference), for women who quitted smoking this happens later at 37 weeks. This might indicate the consequences of smoking withdrawal syndrome which lead to infant growth restriction
• binary logistic regression – backward method for the relationship of preterm birth with the potential risk factors (for periodontitis definition 5 was used) did not result in any model
• the presence of at least 4 pockets did not alter the probability of low birth weight incidence
• principal component analysis-PCA did not result in any relationship of periodontal parameters such as the number of sites with a pocket, the mean pocket depth and the number of sights with bleeding on probing and inflammation and bacteremia markers such as the presence of semi impacted teeth, the presence of destroyed teeth with pulp chamber exposure, the incidence of inflammation of dental etiology during pregnancy and the history of tooth extraction with preeclampsia incidence
Critical review of the literature
• in the literature the relationship between genetic polymorphisms and incidence of low birth weight has not been studied thoroughly
• in the literature genetic polymorphisms of the gene that codes the receptor of vitamin D VDR, rs731236/TaqI and rs7975232/ApaI, appear to be related with both low birth weight and periodontitis. The relationship between polymorphisms rs731236/TaqI and rs7975232/ApaI and low birth weight seems to regard only pregnant mothers of African origin and not pregnant mothers Caucasian heritage. Also the majority of the studies, which report correlation between periodontitis and low birth weight regard pregnant women of African and not Caucasian heritage
• it is possible a part of the incidents of low birth weight infants to occur due to the combined effect of dysbiotic placenta microbiome, which is a result of the hematogenous spread of bacteria from the oral cavity, and the immune response in pregnant women who are genetically predisposed, independently of if there is an active periodontal disease
• there is a need for more studies in order to clarify the possible relationship between maternal periodontitis and low birth weight
